lpress@venera.isi.edu (Laurence I. Press) (09/15/89)
I recently heard Michael Scott-Morton give a talk in which he asserted that U. S. knowledge worker productivity has not been increased by all of our personal computers. I was also told that there was a fairly recent Fortune Magazine article making the same point. 1. Can anyone provide references to research on this question? 2. If this is true, would you care to speculate on why?
eric@snark.uu.net (Eric S. Raymond) (09/16/89)
In <9676@venera.isi.edu> Laurence I. Press wrote: > I recently heard Michael Scott-Morton give a talk in which he > asserted that U. S. knowledge worker productivity has not been increased > by all of our personal computers. I was also told that there > was a fairly recent Fortune Magazine article making the same point. > > 1. Can anyone provide references to research on this question? > > 2. If this is true, would you care to speculate on why? I've been thinking about asking the net that same question myself. It seems to me that the first thing to do would be to investigate whether productivity gains linked to computer use have been recorded elsewhere (i.e. Japan or Europe). It may simply be that our methods of measuring `productivity' are at fault. Is it computed as dollars of corporate revenue generated per employee? What of industries that claim to be computer success stories -- airline reservation systems/travel agencies, accounting firms, banking, etc? Do *they* report hard data on productivity gains? If productivity growth has been flat, what is the limiting factor? Are the machines not powerful enough? Are potential gains lost in the shuffle of procedures designed for paper? Are the increased revenues less than the cost of carrying the systems? There are lots of important questions to be answered here. -- Eric S. Raymond = eric@snark.uu.net (mad mastermind of TMN-Netnews)
al@crucible.UUCP (Al Evans) (09/16/89)
In article <9676@venera.isi.edu> lpress@venera.isi.edu (Laurence I. Press) writes: >I recently heard Michael Scott-Morton give a talk in which he >asserted that U. S. knowledge worker productivity has not been increased >by all of our personal computers. I was also told that there >was a fairly recent Fortune Magazine article making the same point. >[request for references deleted, I don't have any] >2. If this is true, would you care to speculate on why? Wild speculation follows. From a purely experiential viewpoint, I have long suspected that the "productivity gains" wrought by computers are mostly a) ephemeral or b) difficult to define in terms of the "prior art". In my mind, the difficulty lies here: The only really good answer to the non-user question, "What exactly can I DO with a computer?" -- after you've run through the standard spreadsheet-wordprocessor litany -- is "Mostly things you'd never THINK of doing without one." BECAUSE most people and most organizations have gotten along well for years without computers, the possible improvements to be wrought by the adoption of computer assistance are, by definition, outside their experience. To use a musical example, a MIDI studio is not a "better" way of putting together a piece of music -- it's a qualitatively DIFFERENT way. And the music made with it will be different -- unconstrained by certain prior limitations, but constrained by altogether NEW limitations introduced by precisely the technology that makes it possible. The second problem arises when suppositions which seem intuitively correct turn out to be fallacious. A wonderful example of this is the "paperless office," which everybody felt was right around the corner in the early '80s. But instead of being used as a substitute for written documents, computers have become an "efficient" highspeed GENERATOR of same. Another common phenomenon is related to the fact that the computer is great at processing quantitative data, but utterly useless in approaching the qualitative. This leads to the attribution of a higher value to quantitative data -- the "bottom line" -- at the expense of equally important qualitative factors -- customer goodwill, reputation for quality -- which are not susceptible of computerization. I personally know of a case in which cost analyses led to the elimination of such nonessentials as "free beer in the icebox of the employees' lounge". Seems reasonable -- but the elimination of the free beer led DIRECTLY to the loss of several of the company's most productive employees, and the company began to experience financial problems. Even though the people with whom these employess were replaced were, quantitatively, "just as qualified" as the employees who were lost. There is no way (as far as I know) that ANY MIS program could possible equate free beer with enhanced productivity :-) As I see it, then, the problem is primarily one of "trying to put new wine into old skins." We will have to EVOLVE techniques of using computers to increase productivity -- simply doing the old thing faster and more accurately won't work. And we will have to get over our fascination with the reams of precise quantitative data we can produce with 'em, and develop means of giving equal attention to the factors of productivity which cannot, by nature, be analyzed. --Al Evans-- -- Al Evans "You'd grep to know what ...uunet!execu!sequoia!crucible!al you really sed." --Referent Blob
d88-jwa@nada.kth.se (Jon W{tte) (09/17/89)
In <9676@venera.isi.edu> Laurence I. Press wrote: > I recently heard Michael Scott-Morton give a talk in which he > asserted that U. S. knowledge worker productivity has not been increased > by all of our personal computers. I was also told that there > was a fairly recent Fortune Magazine article making the same point. > 2. If this is true, would you care to speculate on why? My little theory on the subject is that what's done on computers is done faster, and/or it looks prettier, but since the computer's so versatile, it's also a wonderful TOY. You can actually look like you're working while you're highly enjoying yourself playing some sort of game (or, you can quickly sontext-switch into a spreadsheet or whatever, especially with the "Set Aside" MF :-) This means, we do as much as we did before, but we do it faster and neater, and thus have more time for recreation while we're being paid for it. Uhh... I'm a consultant, I shouldn't be saying this... But since I work late at nights, I really don't clock my time, but estimates how much time the work done is worth, and my employer's quite happy about that. How many of you folks out there do the same ? :-) -- Life's a bitch, then you die.
arch_ems@gsbacd.uchicago.edu (09/17/89)
>I recently heard Michael Scott-Morton give a talk in which he >asserted that U. S. knowledge worker productivity has not been increased >by all of our personal computers. I was also told that there >was a fairly recent Fortune Magazine article making the same point. > >1. Can anyone provide references to research on this question? > >2. If this is true, would you care to speculate on why? A number of responses have been posted to this query but they all have seemed to skirt an interesting reason... One poster noted that computers allow users to do all sorts of things that couldn't be done before -- another poster noted that 'productivity' was measured in $ of revenue/# of employees. There are two productivity issues which cannot be directly imparted by a $/# calculation. Both have to do with competitiveness and both (_I_ think) are directly affected by computerization: Quality and Complexity. Let me explain -- QUALITY Frankly, I cringe now when I, as a business person, receive a handwritten note of more than five words or a typed resume. I have an expectation that people will use word-processors, and laser printers and desk-top publishing to produce correspondance and important documents like resumes. The standards have gone up regarding the proper public image that an individual or business must present and there is a corresponding increase in cost to produce the higher quality correspondance and publications. If businesses and individuals don't live up to this quality mandate, they will suffer in the marketplace -- images have a lot to do with sales. In addition, there are all sorts of expectations about document delivery. Ten years ago, what was the average cost of mailing things to your business associates? Today, how many fax machines does your business have and how many Federal Express packages do you send a month? (week, day?) This is an increase in the cost of doing business. It directly affects productivity and it is the direct affect of "technology." But again, if you didn't meet this quality expectation, your business would suffer in the marketplace. Finally, there are a lot of expectations regarding the presen- tation of materials now that just didn't apply ten years ago. Graphics, charts, intricate lay out -- all are now a part of the basic documents that not to long ago were pure text. This added quality in the material presentation costs money -- further eroding "productivity." But again -- it is necessary to do these things in order to compete. COMPLEXITY The speed of business transactions and their complexity has expanded as quickly as our increased computing and telecommunications. Many more companies today are operating across international boundaries, in geographically disparate offices, and with highly time sensitive data. Frankly, there are a lot of businesses that are just not possible without the complex infrastructure of technology -- how do you compute that into "productivity?" I mentioned Federal Express before -- how could that business survive without computation? The reason they can charge so much to move a couple of pieces of paper across country for is you is not only because you know it will get their overnight -- but also because if it doesn't you can call them and the person in Nebraska who answers your call from Boston will know that your package to San Francisco is in a delivery truck crossing the Bay Bridge, en route from Oakland (and can also probably tell you that it went to Oakland because you got the zip code wrong, and can tell you the name of the person who will deliver the package and the time within five minutes of when it will be delivered. And I think that is a simple example of the complexity that computers allow -- Now, everyone reading this probably understands intimately the technology that makes all of this possible but you might not have thought about it as part of the bottom line -- there are some other offsets to productivity that are important to consider: training, sunk costs for equipment and networking, system administration, and increased information needs. But the point is -- we are able to do more and more and must in order to remain competitive. --Ted Edward Shelton, Project Manager ARCH Development Corporation arch_ems@gsbacd.uchicago.edu
ck@voa3.UUCP (Chris Kern) (09/18/89)
In article <189@crucible.UUCP> al@crucible.UUCP (Al Evans) writes: >From a purely experiential viewpoint, I have long suspected that the >"productivity gains" wrought by computers are mostly a) ephemeral or >b) difficult to define in terms of the "prior art". > >In my mind, the difficulty lies here: The only really good answer to >the non-user question, "What exactly can I DO with a computer?" -- after >you've run through the standard spreadsheet-wordprocessor litany -- >is "Mostly things you'd never THINK of doing without one." . . . Our experience with office automation tends to confirm this in part. The introduction of a large distributed system that provides, among other things, more-or-less effortless electronic mail and the ability to put shared files on servers where they are easily accessible has reduced the number of telephone calls and memos to some extent. But it has improved internal communication dramatically. The "standard spreadsheet-wordprocessor litany" did help us improve our productivity, especially since a significant segment of our professional staff (radio broadcasters who work in 44 languages) did not even use typewriters before we gave them computer workstations. But the improved communication that our own little computer revolution has brought about has fundamentally changed our organization. -- Chris Kern Voice of America, Washington, D.C. ...uunet!voa3!ck +1 202-485-7020
garton@cunixa.cc.columbia.edu (Bradford Garton) (09/18/89)
In article <189@crucible.UUCP> al@crucible.UUCP (Al Evans) writes: > >As I see it, then, the problem is primarily one of "trying to put >new wine into old skins." We will have to EVOLVE techniques of >using computers to increase productivity -- simply doing the old >thing faster and more accurately won't work. And we will have to >get over our fascination with the reams of precise quantitative >data we can produce with 'em, and develop means of giving equal >attention to the factors of productivity which cannot, by nature, >be analyzed. Very thoughtful answer -- it reminds me a lot of a book I read not too long ago: "In the Age of the Smart Machine" by Harvard sociologist Shoshanna Zuboff. Her thesis is that we are currently undergoing a workplace revolution as far reaching as the Industrial Revolution. To her thinking, our current metrics of "work" and (more importantly) the goals we set for workers based upon those metrics and the *methods* used to achieve those goals must evolve for us to take full advantage of the potential offered by an 'informated' (her term) environment. She looked at a variety of workplace environments for this book ranging from several pulp mills undergoing automation to "high-level" executives at a large banking corporation. One of the wonderful things is that the book has a liberal number of quotations from workers involved with job transformations though computing machinery. The quotes are right on target, and the book raises a number of insightful and thoughtful points. I highly recommend it to anyone interested in this discussion. It even has pictures... Brad Garton Columbia University Music Department brad@woof.columbia.edu
gerwitz@HP-UX.kodak.com (Paul Gerwitz) (09/18/89)
In article <9676@venera.isi.edu> lpress@venera.isi.edu (Laurence I. Press) writes: >I recently heard Michael Scott-Morton give a talk in which he >asserted that U. S. knowledge worker productivity has not been increased >by all of our personal computers. I was also told that there >was a fairly recent Fortune Magazine article making the same point. > >1. Can anyone provide references to research on this question? > >2. If this is true, would you care to speculate on why? I'm not sure if Mr. Morton was responsible for the remarks, but after talking to other management types, I would tend to agree to some extent. It seems that the problem is that most applications in the commercial and manufacturing areas were developed as excelerators of existing paper or human systems. Mike Hammer (MIT, Index Group Prism) aludes to the need for IS to transform, simplify and redeem a business. I recommend an article on Dr. Hammer in Information Week, July 3-10, Page 36. +----------------------------------------------------------------------------+ | Paul F Gerwitz | SMTP: gerwitz@kodak.com | | Eastman Kodak Co | UUCP: ..rutgers!rochester!kodak!eastman!gerwitz | +----------------------------------------------------------------------------+
philba@microsoft.UUCP (Phil Barrett) (09/20/89)
I didn't want to jump in but .... I'm a pretty biased person when it comes to computers but I can describe one specific aspect I have observed that has a strong positive impact on white collar productivity: email. At the `soft', we use email to tremendous benefit. I've worked at companies that didn't have email and I found that dealing with people outside your immediate group to be incredibly inefficient -- phone tag and notes left on peoples desks for example. Now, I just send email. Meetings are an order of magnitude easier to arrange. Another benefit is that it flattens the organization. My boss, his boss, his boss's boss, ... can ask me questions and carry on a conversation with me. Also, you can have discussion groups with out calling meetings. Its not unusual for the chairman of the board to ask questions of individual contributors via email. The entire organization is far more efficient -- decisions are made and information is distributed more quickly. There is a dark side to email. Humor doesn't communicate well. The distribution can be TOO wide and generate controversy/chaos. Its very easy mortally insult someone, with out trying. People can become isolated. Its very easy to ignore mail you don't want to deal with (my biggest sin). Some people just can't write but their verbal expression is good. Even with the bad, I believe that Email connected organizations are much more efficient. I would never want to go back to a non-email environment. Phil Barrett Microsoft I am solely responsible for the above comments
ck@voa3.UUCP (Chris Kern) (09/20/89)
In article <7765@microsoft.UUCP> philba@microsoft.UUCP (Phil Barrett) writes: >Even with the bad, I believe that Email connected organizations are much more >efficient. I would never want to go back to a non-email environment. I'm not sure how many of our 1400 users *could* go back to working without electronic mail. Their dependence on it becomes manifest when one of them experiences a workstation hardware problem. Even though our user support staff typically responds to a call for help in 5 minutes and the vendor (whose field staff is on-site) typically gets a technician to a malfunctioning system in 15 minutes, I often receive urgent requests -- "I had to phone because my computer is down and I can't send mail to you." -- to intercede on a user's behalf to somehow hurry up the repair process. Comments include "I feel completely disconnected" from the rest of the organization and (this usually from a manager whose own workstation is ailing) "I can't get *any* work done." Our electronic mail system is distributed among seven servers; when one goes down, everybody can still send mail and only the users whose "mailboxes" reside on the afflicted server are unable to receive mail. All other system services (local workstation operations such as word processing, use of file servers, etc.) are unaffected. Nevertheless, many people experience the failure of a mail server as a systemic failure. Comments we have heard include "when will you be able to get SNAP (the name of the entire system) up again" and "no electronic mail is getting through"; the latter is patently, and easily provably, false, but it demonstrates the sense of panic many users genuinely feel when there is any failure to meet their expectations about electronic mail. If we tore it all out, we might have to retire a large part of the staff on disability pensions :-). -- Chris Kern Voice of America, Washington, D.C. ...uunet!voa3!ck +1 202-485-7020
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (09/21/89)
In article <7765@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: | At the `soft', we use email to tremendous benefit. I've worked at companies | that didn't have email and I found that dealing with people outside your | immediate group to be incredibly inefficient -- phone tag and notes left | on peoples desks for example. [ more good stuff on email ] You're not alone in that. I am part of a group which installed and is enhancing a corporate email gateway system, between corporate DECnet, local ethernet, internet, usenet, bitnet, QuickComm, etc. We also route for FIDOnet and Compu$erve via gateways. I a recent user survey on services provided, email was rated as "important to productivity" or "very important to productivity" by more respondents than any other service we provide. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
kencr@haddock.ima.isc.com (Kenny Crudup) (09/21/89)
In article <1680@draken.nada.kth.se> d88-jwa@nada.kth.se (Jon W{tte) says: >How many of you folks out there do the same ? :-) What, do stuff like read news on company time? Heck no! No consultants I know! :-) I'm testing NNTP!! -- Kenneth R. Crudup, Contractor, Interactive Systems Co.(386/ix), Cambridge MA Don't worry- I haven't lost my mind....its backed up on tape *somewhere*.... Phone (617) 661 7474 x238 {encore, harvard, spdcc, think}!ima!haddock!kencr kencr@ima.ima.isc.com
dts@quad.uucp (David T. Sandberg) (09/22/89)
In article <7765@microsoft.UUCP> philba@microsoft.UUCP (Phil Barrett) writes: >There is a dark side to email... >Some people just can't write but their verbal expression is good. That sword swings both directions. Some people express themselves much better in writing than they do verbally. I, for one. In that regard, I would rather use email than many other forms of business communication. One factor which I believe contributes to less-than-hoped-for gains from entering the computer realm is that many people/companies are unwilling to embrace the technology to such an extent where they gain the full benefit of it. One place I know of produces a large volume of paper in-house mail, and has ongoing problems with miscommunication. Still, they refuse to use email, even though everyone in the company has a terminal readily available to him/her. I've suggested increased usage of email several times, but they don't want to touch it. -- David Sandberg - Quadric Systems "As of Friday, August 25, 1989, PSEUDO: dts@quad.uucp Triton is a Placemat." ACTUAL: ..uunet!rosevax!sialis!quad!dts
toma@tekgvs.LABS.TEK.COM (Tom Almy) (09/22/89)
In article <425@crdos1.crd.ge.COM> davidsen@crdos1.UUCP (bill davidsen) writes: <In article <7765@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: <| At the `soft', we use email to tremendous benefit. I've worked at companies <| that didn't have email and I found that dealing with people outside your <| immediate group to be incredibly inefficient -- phone tag and notes left <| on peoples desks for example. [ more good stuff on email ] <You're not alone in that. [...] <A recent user survey on <services provided, email was rated as "important to productivity" or <"very important to productivity" by more respondents than any other <service we provide. Yes. But what do you do about computer-phobes that refuse to read their email on a regular basis (or at all), and those people (who are typically important) who demand printed copies of everything? I tried to go paperless back about 1979, to no avail. Tom Almy toma@tekgvs.labs.tek.com Standard Disclaimers Apply "Word Processing" since 1975. Email since 1978 (roughly)
perry@ccssrv.UUCP (Perry Hutchison) (09/23/89)
In article <5978@tekgvs.LABS.TEK.COM> toma@tekgvs.LABS.TEK.COM (Tom Almy) writes: > What do you do about computer-phobes that refuse to read their > email on a regular basis (or at all), and those people (who are > typically important) who demand printed copies of everything? You convince top management to adopt email. When middle managers discover that they are missing messages from the chief by not reading their email, they will quickly learn to do so. They can even learn to print out their own hard-copy if they can't live without it.
ck@voa3.UUCP (Chris Kern) (09/23/89)
In article <5978@tekgvs.LABS.TEK.COM> toma@tekgvs.LABS.TEK.COM (Tom Almy) writes: > ... But what do you do about computer-phobes that refuse to read their >email on a regular basis (or at all) ... Find people fairly senior in each target group who are receptive to using office automation. Those with personal computers at home are often good candidates. Explain, in organizational terms (avoid gee whizzery), what you think the benefits of electronic mail will be. Offer them workstations. Give them an opportunity to discover for themselves how easily they can acquire new information through the use of this informal medium. Their peers will notice that your candidates have better information than they do. Their subordinates will want an opportunity to communicate with them electronically. Sit back and watch the fun. The doubters will be persuaded -- for bureaucratic-political, if not intellectual reasons. It doesn't take long. We reached "critical mass" in a couple of months. After that, there was no going back. Not all our users can articulate the benefits they derive from electornic mail, but we have a strong consensus that it is essential to the management and daily operations of the organization. By the way, be prepared to offer special assistance to those who really are afraid to use computers. Individual training often is a necessity for these users because what really frightens them is the prospect of fumbling while they are learning and consequently looking foolish to their colleagues or subordinates. It often is important to train managers before you train their subordinates. It is also important to restrain the natural tendency that many implementors have to show off all the nifty features of their systems. Figure out what really is important to your new users. Teach them to use those features only. Once they are comfortable using the technology, they will pick up new tricks on their own. And teach you a few, too. -- Chris Kern Voice of America, Washington, D.C. ...uunet!voa3!ck +1 202-485-7020
d88-jwa@nada.kth.se (Jon W{tte) (09/24/89)
In article <5978@tekgvs.LABS.TEK.COM> toma@tekgvs.LABS.TEK.COM (Tom Almy) writes: >Yes. But what do you do about computer-phobes that refuse to read their >email on a regular basis (or at all), and those people (who are typically Send the invitation to the Yearly Company PicNic by email only, show of hands to user@system. >important) who demand printed copies of everything? I tried to go paperless >back about 1979, to no avail. Write a file (typically .impsig) that contains: If you want a printed copy of this message, please type the following three lines at the prompt: s foo !print foo !rm foo :-) h+@nada.kth.se -- $400,000,000,000 is a large sum of money.
roy@phri.UUCP (Roy Smith) (09/24/89)
In article <675@ccssrv.UUCP> perry@ccssrv.UUCP (Perry Hutchison) writes: > You convince top management to adopt email. When middle managers discover > that they are missing messages from the chief by not reading their email, > they will quickly learn to do so. Case in point, I do most of my communication with our Associate Director by email. Unfortunately, he can't seem to deal with reading material on the screen, so he usually takes the documents I email him and prints them out (actually, he eforwards them to his secretary who goes to great pain to strip out the email headers and reformat them for paper!) We have a "computer system oversight" committee here with administrative, scientific, and technical people on it. When organizing meetings, I only send email and make a point of not mentioning it in person, even to people I see on a regular basis, who I suspect may not have bothered to read their email recently. Sometimes it causes people to miss meetings and bitch and moan about it, but I figure it serves them right. -- Roy Smith, Public Health Research Institute 455 First Avenue, New York, NY 10016 {att,philabs,cmcl2,rutgers,hombre}!phri!roy -or- roy@alanine.phri.nyu.edu "The connector is the network"
philba@microsoft.UUCP (Phil Barrett) (09/25/89)
In article <5978@tekgvs.LABS.TEK.COM> toma@tekgvs.LABS.TEK.COM (Tom Almy) writes: > >Yes. But what do you do about computer-phobes that refuse to read their >email on a regular basis (or at all), and those people (who are typically >important) who demand printed copies of everything? I tried to go paperless >back about 1979, to no avail. > Even through this has had some quite reasonable answers, since the question was directed to me, I'll answer it. simple. we just don't hire people that are afraid of computers. I have never heard it said here but I personally believe (the preceding words for the lawyers) computer literacy is an absolute requirement for working in a technology related company such as microsoft. In this company, anybody who demanded printed copies of everything would be weeks behind. That, in fact, was a cornerstone of my orriginal thesis -- decision making occurs at a rapid pace in an E-connected workplace. This brings up an interesting point. In an E-connected environment, there needs to be an easy provision for printing out email. Some people like paper for their files and I like to sometimes print out my mailbox when I get behind (100+ messages per day builds up fast -- one time I came back to 500 pieces of email). You tried this in 1979, the world has come a long way since then. Its probably worth another shot at it. now, what I really want is an email program that is a combination mail manager, database, cross indexer and calandar. The current crop of tools is really running out of gas fast. disclaimer time! yup, you guessed it, the above comments are my opinions and are in no way an official representation of Microsoft corporate policy. Phil Barrett msft
phil@attctc.Dallas.TX.US (Phil Meyer) (09/25/89)
In article <7826@microsoft.UUCP> philba@microsoft.UUCP (Phil Barrett) writes: >In article <5978@tekgvs.LABS.TEK.COM> toma@tekgvs.LABS.TEK.COM (Tom Almy) writes: >> >>Yes. But what do you do about computer-phobes that refuse to read their >>email on a regular basis (or at all), and those people (who are typically >>important) who demand printed copies of everything? I tried to go paperless >>back about 1979, to no avail. >> > >simple. we just don't hire people that are afraid of computers. I have I do systems for Law Firms. email is an integral part of our applications. The problem is this: Professionals view any work done on a keyboard as clerical, and therefore must be done by clerks. This is the single greatest factor in keeping the Legal industry five to six years behind the times. We cannot exclude computer phobes from the interview process. There are NO computer literate legal professionals to speek of! And, of course, ignorance is the basis for most fears. > >now, what I really want is an email program that is a combination >mail manager, database, cross indexer and calandar. The current crop >of tools is really running out of gas fast. Some of my best grey matter is currently tuned to this subject. Especially in the Law Office. Every calendar entry should impact follow-ups, every mail/phone message should impact a calendar or follow-up. If you are working on such a product, I'd like some email. This must become the default in the next generation of office systems. +=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+ | Honesty is in the eye of | Phil Meyer | | the creditor. | Usenet: <backbone>!attctc!lodestar!phil | | | VoiceNet: (214) 991-0897 | +=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+
marksm@syma.sussex.ac.uk (Mark S Madsen) (09/25/89)
In article <425@crdos1.crd.ge.COM> davidsen@crdos1.UUCP (bill davidsen) writes: >In article <7765@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: >| At the `soft', we use email to tremendous benefit. I've worked at companies >| that didn't have email and I found that dealing with people outside your >| immediate group to be incredibly inefficient -- phone tag and notes left > You're not alone in that. I am part of a group which installed and is >enhancing a corporate email gateway system, between corporate DECnet, >...email was rated as "important to productivity" or >"very important to productivity" by more respondents than any other >service we provide. Well, my "corporation" is actually a university, but I take it that scientists also qualify as knowledge workers too, so it is likely that you will be interested to hear my comments. Productivity in research science is largely defined in terms of the number of papers published: this applies to individual researchers, departments, universities, etc. (Don't flame me, I didn't build the system.) I find that email is the single biggest aid to my research. It doesn't enable me to do anything that I couldn't have done without it, except that I can collaborate with someone 5000 miles away at a rate which is more than an order of magnitude higher. Let me demonstrate: WITHOUT EMAIL: Write letters to collaborator; letter cycles take about 10 days. Type up research notes into a paper, mail it to collaborator, wait for reply with suggestions, revise, mail paper to.... Each cycle still takes 10 days at least. The typical time required to agree on a short paper (say, 5 pages) is about 2 months. WITH EMAIL: Send email with notes in LaTeX form. Reply can be received in about 48 hours, allowing for time differences and thinking before replying. Then type paper in LaTeX, email LaTeX source, collaborator inserts changes directly, along with comments in the source. Time taken to write the same paper, about 2 to 3 weeks. Also, some journals now accept submissions (in LaTeX or TeX) via email. Hope this helps the discussion. Mark -- ####################################################################### ## Mark S. Madsen #### marksm@syma.sussex.ac.uk ################### #### Astronomy Centre, University of Sussex, Brighton BN1 9QH, UK. ## #################### Life's a bitch. Then you die. #################
les@chinet.chi.il.us (Leslie Mikesell) (09/25/89)
In article <425@crdos1.crd.ge.COM> davidsen@crdos1.UUCP (bill davidsen) writes: > You're not alone in that. I am part of a group which installed and is >enhancing a corporate email gateway system, between corporate DECnet, >local ethernet, internet, usenet, bitnet, QuickComm, etc. We also route >for FIDOnet and Compu$erve via gateways. Have you dealt with PROFS? Is there any reasonable way to connect it to unix mail? I am currently transferring some files by having the unix machine use a kermit script to log in to VM over a dial-up line so something like PROFS <-> bsmtp might work. Les Mikesell
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (09/26/89)
In article <9662@chinet.chi.il.us>, les@chinet.chi.il.us (Leslie Mikesell) writes: | Have you dealt with PROFS? Is there any reasonable way to connect | it to unix mail? I am currently transferring some files by having | the unix machine use a kermit script to log in to VM over a dial-up | line so something like PROFS <-> bsmtp might work. We have SMTP on the VM side, and Janet on one of the VMS machines. The UNIX<=>VMS is done by (a) SMTP on the VMS machines (some Wolongong some Excelan), and an Ultrix gateway running decnet-ultrix. I would still like to find a source for decnet on UNIX, which could be ported to Suns and Xenix machines. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
peter@ficc.uu.net (Peter da Silva) (09/27/89)
One of the most powerful things a computer can do for an organisation is improve intra- and even inter- office communications enormously. The problem is that most offices computerise by getting a bunch of little personal computers, and maybe a network (though even this is fairly uncommon). So instead of a bunch of people sitting in offices you have a bunch of people and computers sitting in offices. Instead of passing around papers, you pass around floppy disks. Even with a network, the individual computers are single-tasking... a user has to back out of whatever they're doing to send electronic mail. And users have no access to each others' files, either because of network limitations (the network will only work in server *or* client mode on a given box), or because of administrative ones (no or inadequate security). An office is inherently a multi-user environment. Grafting the software on top of a bunch of single-user systems still leaves every man an island. -- Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation. Biz: peter@ficc.uu.net, +1 713 274 5180. Fun: peter@sugar.hackercorp.com. `-_-' "That is not the Usenet tradition, but it's a solidly-entrenched U delusion now." -- brian@ucsd.Edu (Brian Kantor)
peter@ficc.uu.net (Peter da Silva) (09/27/89)
In article <9467@attctc.Dallas.TX.US>, phil@attctc.Dallas.TX.US (Phil Meyer) writes: > The problem is this: Professionals view any work done on a keyboard as > clerical, and therefore must be done by clerks. This is the single greatest > factor in keeping the Legal industry five to six years behind the times. This is pretty amazing. The first commercial application for UNIX I ever heard of was office automation in a law firm. Email was prominently mentioned in the article, which I read in 1978 from an aged photocopy, so it was probably 1977 or earlier. Was this a fluke? Anyone have more information on it? -- Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation. Biz: peter@ficc.uu.net, +1 713 274 5180. Fun: peter@sugar.hackercorp.com. `-_-' "That is not the Usenet tradition, but it's a solidly-entrenched U delusion now." -- brian@ucsd.Edu (Brian Kantor)
ck@voa3.UUCP (Chris Kern) (09/27/89)
In article <6313@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: >[...] Even with a network, the individual computers are single-tasking... a >user has to back out of whatever they're doing to send electronic mail. And >users have no access to each others' files, either because of network >limitations (the network will only work in server *or* client mode on a >given box), or because of administrative ones (no or inadequate security). > >An office is inherently a multi-user environment. Grafting the software >on top of a bunch of single-user systems still leaves every man an island. The point about multitasking is well-taken, but that doesn't necessarily imply that a shared-logic machine is the only approach to office automation. Single-user multitasking workstations are an alternative and sometimes superior way to deliver computer resources to professionals. One of the implementor's most important tasks is to match the technology to the organization. Most of our users have multitasking workstations. A few use IBM-ish personal computers, which are connected to the same network. The users with the workstations tend to think of the monitors in front of them as windows into the organization. Electronic mail and shared mass storage have displaced and significantly augmented traditional forms of communication. The users with the MS-DOS machines have a radically different view of the technology. As Peter suggests, they essentially use personal computers that have been taught a few new tricks. They might as well be connecting to a dial-up port. The network buys them nothing except a high transfer rate. To really do the job properly, however, most operations -- not just notification and retrieval of electronic mail -- should take place in the background (or at least the user should have that option). One of the things that distinguishes professionals from clerical or mechanical workers is that the professionals are expected to manage their own time. Typically, this means that work cannot be arranged serially; usually, the professional has to be able to do several things at once. The first release of software that we installed on our workstations often blocked the employee from using the machine while a network operation (electronic mail retrieval or transmission, storage or retrieval on a file server, etc.) was taking place. We have a much lower level of user frustration now that almost all workstation functions either take place automatically in the background or give the user the option to select foreground (for quick completion) or background operation. Providing proper distribution of network resources is also an issue, but using small single-function servers works well in our environment. Except in rare instances, none of our workstations acts as a server to another system. There are cases where one *server* will act as a client of another server -- authentication sometimes requires an operation of this kind -- but as long as the servers are also multitasking, that isn't an inherent problem. Finally, needless to say authentication (security) is important, even if your application doesn't involve really secret stuff. But the problem is not insoluble. See the work on Kerberos at MIT or, for a mature implementation, see the authentication service that is included in Xerox Clearinghouse Service. -- Chris Kern Voice of America, Washington, D.C. ...uunet!voa3!ck +1 202-485-7020
arch_ems@gsbacd.uchicago.edu (09/27/89)
>One of the most powerful things a computer can do for an organisation >is improve intra- and even inter- office communications enormously. The >problem is that most offices computerise by getting a bunch of little >personal computers, and maybe a network (though even this is fairly >uncommon). BUT becoming more and more common. Especially in the MAC world where every machine comes with built in networking hardware/software. A recent (August 1, 1989) MacWEEK "Marketwatch" report states: "...24 percent of (surveyed) sites have Macintoshes connected to LANs at other locations.... 93 percent have Macintoshes connected to local LANs..." Now, they only surveyed big (to me) companies with hundreds of machines but ut is an interesting figure. Here at the University of Chicago, most machines in departments are on LANs and many LANs are connected to the University WAN... >Even with a network, the individual computers are single-tasking... a user >has to back out of whatever they're doing to send electronic mail. And users >have no access to each others' files, either because of network limitations >(the network will only work in server *or* client mode on a given box), or >because of administrative ones (no or inadequate security). Poor guy, you must be in the IBM - PC (clone) world because all of these things are not problems in my MAC universe. Right now I have TELNET running to connect me to a VAX (via ethernet) wich runs my news reader, a window under it with my office email system running (CE Software's QuickMail), a wordprocessing program under that... you get the picture. we have adequate security... file sharing... etc. QuickMail is actually being served as a background process on my machine to the other 8 macs in my office and (via an RSERVER modem bridge) to a number of people working from their homes and to another office across campus (with its 17 machines)... > >An office is inherently a multi-user environment. Grafting the software >on top of a bunch of single-user systems still leaves every man an island. I've come up with a new slogan to rouse PC people -- "Rejoin the technological Revolution!" -:) Hey, proof -- I just took a message for someone who hasn't come into the office yet -- in the middle of writing this response I flipped to my QuickMail screen, typed in the phone message, shot it off to her, and flipped back to typing this response... Edward Shelton, Project Manager ARCH Development Corporation arch_ems@gsbacd.uchicago.edu
peter@ficc.uu.net (Peter da Silva) (09/28/89)
Let me clarify one thing... In article <294@voa3.UUCP>, ck@voa3.UUCP (Chris Kern) writes: > In article <6313@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: > >An office is inherently a multi-user environment. Grafting the software > >on top of a bunch of single-user systems still leaves every man an island. > The point about multitasking is well-taken, but that doesn't necessarily > imply that a shared-logic machine is the only approach to office automation. I defined multuser in functional terms: multitasking and protection. There is nothing there forcing these multiuser machines to actually have more than 1 user on them. Multiuser is a function of the software on the machine, not the distribution of CPU power. The rest of the article basically reinforces my point. -- Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation. Biz: peter@ficc.uu.net, +1 713 274 5180. Fun: peter@sugar.hackercorp.com. `-_-' "That is not the Usenet tradition, but it's a solidly-entrenched U delusion now." -- brian@ucsd.Edu (Brian Kantor)
jeffd@ficc.uu.net (jeff daiell) (09/28/89)
In article <6331@ficc.uu.net>, peter@ficc.uu.net (Peter da Silva) writes: > Let me clarify one thing... > > In article <294@voa3.UUCP>, ck@voa3.UUCP (Chris Kern) writes: [Chris' comments and Peter's response deleted] > The rest of the article basically reinforces my point. Wait a sec, Peter. When we were hashing out The Market vs. The State in tx.politics several weeks ago, you would *never* have let me get away with a statement like the above. So please explain *how* the rest of the article basically reinforces your point. Jeff Daiell -- "Truth has always been found to promote the best interests of mankind..." -- Percy Bysshe Shelley
les@chinet.chi.il.us (Leslie Mikesell) (09/29/89)
In article <6313@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: >Even with a network, the individual computers are single-tasking... a user >has to back out of whatever they're doing to send electronic mail. And users >have no access to each others' files, either because of network limitations >(the network will only work in server *or* client mode on a given box), or >because of administrative ones (no or inadequate security). It doesn't *have* to be that way. We use a pop-up mailer that works the same whether you are on the network or dialing up (in the background) so you don't have to stop what you are doing on the PC to answer mail. It can also handle attached files, so there is no problem with sharing, although on the network a common area on a server is handier. >An office is inherently a multi-user environment. Grafting the software >on top of a bunch of single-user systems still leaves every man an island. No, it just takes a little thought about where the data should be stored. Most network-aware PC software knows as much or more about protecting shared files during simultaneous access than most unix programs. Les Mikesell
philba@microsoft.UUCP (Phil Barrett) (09/29/89)
In article <6313@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: >One of the most powerful things a computer can do for an organisation >is improve intra- and even inter- office communications enormously. The >problem is that most offices computerise by getting a bunch of little >personal computers, and maybe a network (though even this is fairly >uncommon). So instead of a bunch of people sitting in offices you have >a bunch of people and computers sitting in offices. Instead of passing >around papers, you pass around floppy disks. > >Even with a network, the individual computers are single-tasking... a user >has to back out of whatever they're doing to send electronic mail. And users >have no access to each others' files, either because of network limitations >(the network will only work in server *or* client mode on a given box), or >because of administrative ones (no or inadequate security). > >An office is inherently a multi-user environment. Grafting the software >on top of a bunch of single-user systems still leaves every man an island. Hmmm, let me see. Yeah, I'm running on a compaq. Yup, there's dos down there somewhere. Yup, excel, my bug database and mail program are all still running. Not to mention my vtp to the local news server which is what I'm using to post this message (at least when I'm not switching back to make sure the other stuff is running . The network seems to still be up -- and so are the servers I'm connected to. And I could run this stuff on a lowly 286. Maybe you haven't looked at what's around lately but there are numerous email packages commercially available. DaVinci has one for Windows and there are numerous DOS versions that can be multitasked under windows, os/2 or desqview. A lot of fortune 1000 companies do this now. PC networking is becomming a very big business and a *lot* more common than you think, read PC Mag, PC WORLD, etc. I don't know the actual sales but they are enough to attract numerous large players. Microsoft, Novell, IBM, 3Com, UB, TOPS, 10Net (DCA?), etc all make reasonable products. So, quite simply, I can put together a very nice office automation system with file servers, shared printers, Email and PCs AND still get all the benefits of all that PC SW out there. Also, there are lots of DOS based SW products designed to work nicely on a network. You are right that the DOS based networks dont support both server and client on a single machine reasonably. But so what? File sharing is quite easy; I don't remember the last time I passed a diskette around. Security isn't super tight but its good enough for many (if not most) installations. I've used multiuser systems for years and I really would never want to go back to one. The amount of SW available is so incredibly limited compared to PCs. I certainly dont feel like an island, not with >75 email messages a day. Phil `the above opinions are mine' barrett Microsoft
peter@ficc.uu.net (Peter da Silva) (09/29/89)
In article <7886@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: > Hmmm, let me see. Yeah, I'm running on a compaq. Yup, there's dos down > there somewhere. Yup, excel, my bug database and mail program are > all still running. Not to mention my vtp to the local news server which > is what I'm using to post this message (at least when I'm not switching > back to make sure the other stuff is running . The network seems to still > be up -- and so are the servers I'm connected to. And I could run this > stuff on a lowly 286. Yeh, with no end of FM in there to make it work right. Back when MS-DOS was fairly new, I was doing that sort of stuff on a lowly PC-XT running your company's own Version-7 based Microsoft Xenix. Yes, an XT with a 20 meg drive. Of course, these days UNIX is a lot bigger... but it's also a whole lot more powerful. And everything works without the FM. If Bill Gates had any vision left, MS-DOS 3.0 *would* have been Xenix. And of course everythings co-operating now, but remember when you were putting it together. I remember the sorts of things you have to do: "Now, if I run DoubleDOS before Sidekick then I can't get to Sidekick in this window, but if I do it the other way then this doesn't work. How about...". With a real operating system (UNIX, for example, or one of the proprietary operating systems you've plowed under with IBM's money), everything just plain works. > So, quite simply, I can put together a very nice office automation system > with file servers, shared printers, Email and PCs AND still get all the benefits > of all that PC SW out there. Also, there are lots of DOS based SW products > designed to work nicely on a network. That last sentence is the critical one here. You have to have all your applications especially coded to work with the network. I'm still running UNIX software that was born before Ethernet. It hasn't been updated. It doesn't have to be... it doesn't know the nets there and doesn't need to. > You are right that the DOS based networks dont support both server and client > on a single machine reasonably. But so what? Transparent file access is like sex. If you haven't tried it, you don't miss it. If you have tried it, yu can't get along without it. > File sharing is quite easy; > I don't remember the last time I passed a diskette around. i.e., you use the network as a fast modem port. > Security isn't > super tight but its good enough for many (if not most) installations. Had a look in comp.virus recently? > I've > used multiuser systems for years and I really would never want to go back to > one. The amount of SW available is so incredibly limited compared to PCs. If you don't mind having to bring your car to a halt before you can turn the steering wheel, more power to you. Besides, I can already run all that dos software. Ever hear of emulators? And I don't have to worry about program A treading on program B's toes. -- Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation. Biz: peter@ficc.uu.net, +1 713 274 5180. Fun: peter@sugar.hackercorp.com. `-_-' "That is not the Usenet tradition, but it's a solidly-entrenched U delusion now." -- brian@ucsd.Edu (Brian Kantor)
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (09/30/89)
Someone asked how you get people to use email. The secret is to have support at the top. If the head honcho starts asking for info by email, sending out meeting notices, etc, pretty soon the next level down starts at least reading and responding. The technical staff pretty much will use it as soon as available, anyway, and the middle management gets mail from above and below. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
campbell@redsox.bsw.com (Larry Campbell) (09/30/89)
In article <7886@microsoft.UUCP> philba@microsoft.UUCP (Phil Barrett) writes:
-Hmmm, let me see. Yeah, I'm running on a compaq. Yup, there's dos down
-there somewhere. Yup, excel, my bug database and mail program are
-all still running. Not to mention my vtp to the local news server which
-is what I'm using to post this message (at least when I'm not switching
-back to make sure the other stuff is running . The network seems to still
-be up -- and so are the servers I'm connected to. And I could run this
-stuff on a lowly 286.
And what multitasking operating system are you running? Oh, you're not?
You're running DOS? With a bunch of TSR hacks loaded? Oh, OK, how would
one go about writing such a TSR, reliably? How do I know it's not going to
step on someone else's TSR? I just follow a bunch of undocumented rules,
right? Wouldn't it be nice if these rules were codified, standardized, and
documented? Wouldn't people like Borland (Sidekick), Lotus (Metro), and the
Prokey folks love to have a _reliable_ way to write TSRs without crashing
your PC? Didn't these people talk Microsoft into participating in such a
standardization effort? And didn't Microsoft tube the whole project in
order to try to shove OS/2 down our throats instead?
Pardon me while I fetch my airsick bag...
--
Larry Campbell The Boston Software Works, Inc.
campbell@bsw.com 120 Fulton Street
wjh12!redsox!campbell Boston, MA 02146philba@microsoft.UUCP (Phil Barrett) (10/02/89)
In article <6361@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: > >And of course everythings co-operating now, but remember when you were >putting it together. I remember the sorts of things you have to do: "Now, >if I run DoubleDOS before Sidekick then I can't get to Sidekick in this >window, but if I do it the other way then this doesn't work. How about...". I'm not sure what your point is. Are you saying that systems integration under a DOS environment is impossible or so difficult as require a guru to get it working? If so, Patently false. Yes, it takes some knowledge but what computer systems dont? Are you saying that there are problems in systems integration for any environment? Then I have to agree. I've worked in several environments and put together dos, proprietary and unix systems and none were particularly easier than the others. I know of numerous small shops that use networking and PCs and other than a day or two of the `Reps' time, they all went together pretty easily and work. One shop is a medical benefits plan administration group. They track invoices, claims, new clients, etc with 15 PCs and Novell netware. They use a standard database (rBase, I think) multiuser/networked version, ms excel, ms word, some mailing list SW and a few others. It works and they are happy. They just added several PCs and it took more time pulling the PCs out of the boxes and getting them running than it did to get them on the network and using the database. Not a guru in sight (or on site). > >With a real operating system (UNIX, for example, or one of the proprietary >operating systems you've plowed under with IBM's money), everything just >plain works. I guess we just plain disagree. By the way, don't confuse the person with the company. The statement -- everything just plain works is maybe true for a limited set of applications but I find it odd coming from a *NIX support guy. How many times has uucp gotten wedged on you? How many times have you gotten some SW and discovered it hadn't been implemented for or tested on your flavor of *NIX? Don't get me wrong, Unix is great and it has many nice features over DOS but its by no means perfect. There a lots of problems getting a *NIX up and running with all the applications you need. There is no question in my mind that DOS could be a *lot* better but the depth and breadth of applications for it make it a compelling choice to most of the microcomputer users. Binary Compatibility is a key factor -- one which the *NIX market sadly lacks. Its not at all clear the Emulators under unix work that well. There are numerous types of apps that don't work at all and it varys from system to system. > >> So, quite simply, I can put together a very nice office automation system >> with file servers, shared printers, Email and PCs AND still get all the benefits >> of all that PC SW out there. Also, there are lots of DOS based SW products >> designed to work nicely on a network. > >That last sentence is the critical one here. You have to have all your >applications especially coded to work with the network. I'm still running >UNIX software that was born before Ethernet. It hasn't been updated. It >doesn't have to be... it doesn't know the nets there and doesn't need to. Ahem... Perhaps I should have been clearer. Given PC-NET, MS LanMan or Novell nets, file access IS transparent. No special `coding' is required. Other than my multiuser bug database, I use no network aware SW and it all works over the net just fine. Multi-user databases management systems, etc DO require special network support. I don't see this as significantly different from the sort of things one has to do for a DBMS, etc in a multiuser environment. (record/file locking, for example) > >> You are right that the DOS based networks dont support both server and client >> on a single machine reasonably. But so what? > >Transparent file access is like sex. If you haven't tried it, you don't >miss it. If you have tried it, yu can't get along without it. see above. its available with all the major PC networks. You simply use a server based file model rather than one on your personal machine. There are several large banks that use this approach and they are very happy. > >> File sharing is quite easy; >> I don't remember the last time I passed a diskette around. > >i.e., you use the network as a fast modem port. I could see how you would think this but its simply not true. Many of our shared files reside on a server and are accessed directly. The aformentioned medical benefits shop keeps their database on a server. The banks above keep most of their shared data on a server. > >> Security isn't >> super tight but its good enough for many (if not most) installations. > >Had a look in comp.virus recently? seen any good internet viruses lately? Found any good security holes in unix lately? Point is that no system is totally secure. Besides, if viruses are a concern, there are simple methods of avoiding them (like don't run stuff off BBSs and don't boot off of suspected diskettes and ...) as well as programs like flu-shot and such. > >> I've >> used multiuser systems for years and I really would never want to go back to >> one. The amount of SW available is so incredibly limited compared to PCs. > >If you don't mind having to bring your car to a halt before you can turn >the steering wheel, more power to you. sigh... I think you still haven't gotten it. I can multitask dos apps to my hearts content. There are numerous solutions that work: MS-Windows, os/2, DeskView, SW Carosel (sp?), VM386, ... Not only do I not need to stop the car, I can drive several at once and I can cut and paste between dos apps with several of the above products. Hundreds of thousands of PC users use these products daily so there must something wrong :-) I realize that this won't change your mind. All I'm trying to say is that the statement that PCs are unusable for anything other than a single task at a time in an isolated environment is simply wrong. No environment is perfect, it is usability that counts. You can argue the details but a lot of people use these kind of systems to significant benefit every day. Phil Barrett Microsoft The above opinions are mine and do not necessarily reflect those of my employer.
peter@ficc.uu.net (Peter da Silva) (10/02/89)
In article <7916@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: > I'm not sure what your point is. I'll spell it out. Systems integration under DOS, for anything but "load one program at a time and run it", is prohibitively difficult for the vast majority of users, who are not technically oriented. It doesn't need a guru, but it's far more difficult than it needs to be. > I guess we just plain disagree. By the way, don't confuse the person with > the company. The statement -- everything just plain works is maybe true for > a limited set of applications but I find it odd coming from a *NIX support > guy. How many times has uucp gotten wedged on you? Under a modern system-V system? Never. How many times has Fido gotten wedged on you. Oh, that's right. Fido just shuts down at 4AM every day so it doesn't have to deal with the problem. > How many times have you > gotten some SW and discovered it hadn't been implemented for or tested on > your flavor of *NIX? A couple of times for our old system-III based Xenix-286 boxes, but that's to be expected. It's like expecting your DOS software to run on MS-DOS 1.4 on a Victor-9000. Under System-V, never. > Binary Compatibility is a key factor -- one which the *NIX > market sadly lacks. Can you say "System-V on an 80386 with an AT-bus"? I've even got binary device drivers to port. Application support is flawless. But if you need more power than a microcomputer, you have a route all the way to big iron. > Its not at all clear the Emulators under unix work > that well. DOS emulators on said System-V/386 box have been very good to me. > >> Security isn't > >> super tight but its good enough for many (if not most) installations. > >Had a look in comp.virus recently? > seen any good internet viruses lately? That's a red herring. The internet is deliberately designed as a low- or no- security environment. And for all that it's more secure than any PC net. I'm on the security mailing list, and have yet to see a System-V related problem show up. > Found any good security holes in > unix lately? In BSD maybe. Not under System-V. > Point is that no system is totally secure. True, but running DOS is like leaving your door unlocked as a matter of course. Yes, a good burglar can still get in anywhere, but you don't need to encourage them. > Besides, if viruses > are a concern, there are simple methods of avoiding them (like don't run stuff > off BBSs and don't boot off of suspected diskettes and ...) ... and don't buy shrink-wrap software from a computer store, and so on. You *haven't* been reading comp.virus, have you? > I think you still haven't gotten it. I can multitask dos apps to my hearts > content. I've been there. I tried just about everything, and the only system that could support more than one badly-behaved application in anything like a reliable manner was DoubleDOS. And I still kept blowing away directories when I accidentally tried to access the one file from both partitions. Windows, DesqView, and the rest were far worse. > Hundreds of thousands of PC users use these products > daily so there must something wrong :-) Yes, I always use that as an argument why you shouldn't assume the majority is always right. > I realize that this won't change your mind. All I'm trying to say is that > the statement that PCs are unusable for anything other than a single task > at a time in an isolated environment is simply wrong. I don't think I ever used the term "unusable". Kludge, yes, and baroque, yes, and a waste of resources, yes. But not unusable. Hell, I've even recommended the bloody things. But setting up an office around a bunch of networked DOS machines is a dead end. And the cancerous spread of the beasts has hurt productivity. -- Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation. Biz: peter@ficc.uu.net, +1 713 274 5180. Fun: peter@sugar.hackercorp.com. `-_-' "That is not the Usenet tradition, but it's a solidly-entrenched U delusion now." -- brian@ucsd.Edu (Brian Kantor)
bill@twwells.com (T. William Wells) (10/02/89)
In article <6375@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes:
: > >Had a look in comp.virus recently?
:
: > seen any good internet viruses lately?
:
: That's a red herring. The internet is deliberately designed as a low- or no-
: security environment. And for all that it's more secure than any PC net.
: I'm on the security mailing list, and have yet to see a System-V related
: problem show up.
Really?
What about the one that lets any user trivially truncate the password
file? And lets the real clever get root access?
---
Bill { uunet | novavax | ankh | sunvice } !twwells!bill
bill@twwells.comsl@van-bc.UUCP (Stuart Lynne) (10/03/89)
In article <6375@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: }In article <7916@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: }> I'm not sure what your point is. } }> a limited set of applications but I find it odd coming from a *NIX support }> guy. How many times has uucp gotten wedged on you? } }Under a modern system-V system? Never. I'll second that. Van-bc logs 200+ hours a week of uucp traffic on four lines. We have *NEVER* had uucp get stuck since we installed the 386 with SCO last winter. We have had minor problems finding a multi-port serial card that would work with modem control. But that's not a Unix problem. And the SCO Async driver has always functioned flawlessly on the builtin serial ports. -- Stuart.Lynne@wimsey.bc.ca uunet!van-bc!sl 604-937-7532(voice) 604-939-4768(fax)
palowoda@fiver.UUCP (Bob Palowoda) (10/03/89)
From article <7916@microsoft.UUCP>, by philba@microsoft.UUCP (Phil Barrett): > In article <6361@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: >> [some stuff deleted] > microcomputer users. Binary Compatibility is a key factor -- one which the *NIX > market sadly lacks. With respect to microcomputers Binary Compatibility is *NOT* a key factor. I run Xenix286/886, Microport, ESIX, ATT, ISC's binary applications on my SysV machine with no major problems. When I have found problems they where usally 286 bin's compiled with the goofy tunes MSC compiler causeing wierd pionters trying to write where it shouldn't. Buy that's just my opinion. > Its not at all clear the Emulators under unix work > that well. There are numerous types of apps that don't work at all and it > varys from system to system. I use Simul-Task 2.0 (VPIX) and to many programs work to pass up. Excuse me too many good apps that I need to complete my work. >> >>> So, quite simply, I can put together a very nice office automation system >>> with file servers, shared printers, Email and PCs AND still get all the benefits >>> of all that PC SW out there. Also, there are lots of DOS based SW products >>> designed to work nicely on a network. >> >>That last sentence is the critical one here. You have to have all your >>applications especially coded to work with the network. I'm still running >>UNIX software that was born before Ethernet. It hasn't been updated. It >>doesn't have to be... it doesn't know the nets there and doesn't need to. > > Ahem... Perhaps I should have been clearer. Given PC-NET, MS LanMan or > Novell nets, file access IS transparent. No special `coding' is required. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ What do you mean by this? [cut cut] >> >>> You are right that the DOS based networks dont support both server and client >>> on a single machine reasonably. But so what? >> >>Transparent file access is like sex. If you haven't tried it, you don't >>miss it. If you have tried it, yu can't get along without it. > > see above. its available with all the major PC networks. You simply use > a server based file model rather than one on your personal machine. I used to think this was a good idea, now I think it's a shame that someone spends all that money for a dedicated machine just to make it a fileserver. Really it's a waste of expensive hardware. But DOS users don't have many alternatives. I was setting up realestate office's with some Novell networks 10 to 15 units per office when I lost a contract to someone because the owner was convenced that NFS could run as clients and servers with the same database he was useing on dos. >> >>> Security isn't >>> super tight but its good enough for many (if not most) installations. >> >>Had a look in comp.virus recently? > > seen any good internet viruses lately? Found any good security holes in > unix lately? Point is that no system is totally secure. Besides, if viruses > are a concern, there are simple methods of avoiding them (like don't run stuff > off BBSs and don't boot off of suspected diskettes and ...) as well as > programs like flu-shot and such. Egad, thier talking about commercial software for dos containing viruses. If you take your above approach you shouldn't run any software to be secure. Speaking of dos viruses I got a hold of one to play around with. Sence the emulator I use for dos creates a drive C as nothing but a pseudo file the stupied virus attached itself to the file. No big deal just reinstall dos. All my dos apps where in the unix parition anyways. > > I realize that this won't change your mind. All I'm trying to say is that > the statement that PCs are unusable for anything other than a single task > at a time in an isolated environment is simply wrong. No environment is > perfect, it is usability that counts. You can argue the details but a lot of > people use these kind of systems to significant benefit every day. I will not disagree with you here. But what seems to be happening is that developers are running into some limitations writeing software in multiuser DOS machines. This kind of puts a halt on thier advancement to keep up on the competition. OS2 may be the answer but it seems it it's rought stages right now. If a developer gets stuck by any of these limitation thier going to try alternatives. That dosn't mean they are not going to develop OS2 apps. They just have a little more time to develop on two OS's. When MS gets the bugs worked out or enhances the the OS to the point the developer can get thier apps working they way they want it they release to OS2 version. Your right "usabilty" is what counts. But what comes after that is "reliablity". Both terms are too subjective. However the costomers money isn't. ---Bob -- Bob Palowoda packbell!indetech!palowoda *Home of Fiver BBS* login: bbs Home {sun|daisy}!ys2!fiver!palowoda (415)-623-8809 1200/2400 Work {sun|pyramid|decwrl}!megatest!palowoda (415)-623-8806 2400/9600/19200 TB Voice: (415)-623-7495 Public access UNIX XBBS
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/03/89)
In article <7916@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: | I guess we just plain disagree. By the way, don't confuse the person with | the company. The statement -- everything just plain works is maybe true for | a limited set of applications but I find it odd coming from a *NIX support | guy. How many times has uucp gotten wedged on you? On SysV? It doesn't happen. And if you applied the patches not on SysIII either. V7 was bad, but that's no more realistic than mentioning Dos 2.0 bugs today. | How many times have you | gotten some SW and discovered it hadn't been implemented for or tested on | your flavor of *NIX? There used to be a lot of net stuff which was BSD only, but the people who wrote it are now finding that portability is important, and most runs on SysV. Commercial software of any quality runs BSD and SysV. I suppose you could find some that doesn't, but then you can find DOS software which doesn't work, too. And *dozens* of TRS which all say "load me last." Portability isn't the problem it was even a few years ago in the UNIX world, while DOS has become less stable. Windows, Desqview, DoubleDOS, NetWare... all have things which don't run under the other, and all have some TSR's which mess them up. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
peter@ficc.uu.net (Peter da Silva) (10/04/89)
Our friend at Microsoft's position seems to be that things are much better in the DOS world than I think. Well, things might have improved in the past two years (last time I was extensively DOSing), but that still doesn't make up for the isolation among DOS users for the 7 years prior to that. Nor for the continued low productivity of knowledge workers at sites that don't get the latest and greatest. The first office automation system using UNIX was in something like 1976. MP/M and Cromix were providing multiuser CP/M systems in 1980. By 1989 DOS seems to be catching up, given the claims of transparent file sharing on DOS networks. Just think were we'd be today if we hadn't taken a ten year time-out. -- Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation. Biz: peter@ficc.uu.net, +1 713 274 5180. Fun: peter@sugar.hackercorp.com. `-_-' ``I feel that any [environment] with users in it is "adverse".'' 'U` -- Eric Peterson <lcc.eric@seas.ucla.edu>
philba@microsoft.UUCP (Phil Barrett) (10/06/89)
In article <6375@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: >In article <7916@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: >> I'm not sure what your point is. > >I'll spell it out. Systems integration under DOS, for anything but "load one >program at a time and run it", is prohibitively difficult for the vast >majority of users, who are not technically oriented. It doesn't need a >guru, but it's far more difficult than it needs to be. and not for unix? Desqview and windows have very user friendly installation programs. > >> I guess we just plain disagree. By the way, don't confuse the person with >> the company. The statement -- everything just plain works is maybe true for >> a limited set of applications but I find it odd coming from a *NIX support >> guy. How many times has uucp gotten wedged on you? > >Under a modern system-V system? Never. wow! thats pretty amazing. maybe unix *is* approaching reliability. > >How many times has Fido gotten wedged on you. Oh, that's right. Fido just >shuts down at 4AM every day so it doesn't have to deal with the problem. who uses fido? I certainly don't. I don't know of any sites that do. But since Fido has nasty problems, I guess that means all PC comm apps do too :-) > >> Binary Compatibility is a key factor -- one which the *NIX >> market sadly lacks. > >Can you say "System-V on an 80386 with an AT-bus"? I've even got binary device >drivers to port. Application support is flawless. flawless? wow, a perfect system. amazing! the unix market is fragmented. there are a miriad of architectures that it runs on. Thats its strength. But its also a weakness. try running a SystemV 386 app on a sun 3. There are something like 40M machines that can run DOS apps. Whats the installed base of any version of Unix? > >> Its not at all clear the Emulators under unix work >> that well. > >DOS emulators on said System-V/386 box have been very good to me. do they support mice well? windows apps? graphical apps? TSRs? `dirty' apps, dos extended apps? cut and paste between apps? > >> >> Security isn't >> >> super tight but its good enough for many (if not most) installations. > >> >Had a look in comp.virus recently? > >> seen any good internet viruses lately? > >That's a red herring. The internet is deliberately designed as a low- or no- >security environment. And for all that it's more secure than any PC net. >I'm on the security mailing list, and have yet to see a System-V related >problem show up. Its not a red herring. Just because it was designed that way doesn't mean squat. security is only as good as the weakest link. The PC networks do have some limited security features -- server password protection. Granted that this isn't a complete solution but the environment is much more limited than say arpa or uucp. > >> Besides, if viruses >> are a concern, there are simple methods of avoiding them (like don't run stuff >> off BBSs and don't boot off of suspected diskettes and ...) > >... and don't buy shrink-wrap software from a computer store, and so on. You >*haven't* been reading comp.virus, have you? Oh yeah, every package you buy has at least 3. :-) I especially like the one where someone claimed to have bought a package with some large number of viruses infecting the single package. Most of these said virus laden packages are totally unsubstantiated. A lot of them have come from SW rental stores where they have a shrink wrapper in the back room. Who knows where the disks have been. There have been several real cases, though. The FreeHand one was probably the most well publicised. The people at Aldus are totally hard core about viruses now -- I'd be real suprised to see something like that from them again. Most companies are *very* careful. We take carefull steps here to prevent this happening to us. No SW company needs that rap. And yes, I read comp.virus. > >I don't think I ever used the term "unusable". Kludge, yes, and baroque, yes, >and a waste of resources, yes. But not unusable. Hell, I've even recommended >the bloody things. But setting up an office around a bunch of networked >DOS machines is a dead end. And the cancerous spread of the beasts has >hurt productivity. You are right, you didn't say unusable, just about everything but, though. You did say that they were isolated environments where data sharing is done by passing diskettes around -- which I showed via example is not true at all. I grant that Unix is far more elegant in most ways but elegance isn't everything. You may have disgust for the majority but people do vote with their $$$. Mainframe systems, Unix, VAXes, etc simply don't solve their problems -- if they could, why hasn't there been an enormous rush to buy these products for office automation. Thats not to say that there isn't a need for those machines but look at the market researce, mainstream corporate computing uses networked PCs. Well, this thread isn't going anywhere but its been fun. Have a nice night! Phil And of course, the above opinions are mine.
utoddl@uncecs.edu (Todd M. Lewis) (10/06/89)
In article <6375@ficc.uu.net>, peter@ficc.uu.net (Peter da Silva) writes: > I don't think I ever used the term "unusable". Kludge, yes, and baroque, yes, > and a waste of resources, yes. But not unusable. Hell, I've even recommended > the bloody things. But setting up an office around a bunch of networked > DOS machines is a dead end. And the cancerous spread of the beasts has > hurt productivity. > -- > Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation. Gee, Peter. You have this annoying habit of being painfully right. And there you go again, cutting right to the bone and exposing the gleaming red of bloody truth. I've spent the last five years setting up PCs, networks, installing applications, and electronically reorganizing offices all over our campus, and every day it has become more clear that _there_has_to_be_a_better_way_. Sure, DOS works, and you can do things with it, but it is a dead end. If the effort that has been put into making so much software work on such a limited OS had been put into a real, mature OS, we would all be much better off than we are now. 10 million lemmings can't be wrong. Sure. _____ | Todd M. Lewis Disclaimer: If you want my employer's ||\/| utoddl@ecsvax.uncecs.edu ideas, you'll have to || || utoddl@ecsvax.bitnet _buy_ them. | || |___ ("Prgrms wtht cmmnts r lk sntncs wtht vwls." --TML)
peter@ficc.uu.net (Peter da Silva) (10/06/89)
In article <7971@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: > and not for unix? Desqview and windows have very user friendly installation > programs. That's as may be, but that's not what I mean by "system integration". The problem with DOS comes when you have <n> different programs all with their own kludges put in to cover up the faults of MS-DOS (like, the serial port support just plain doesn't work). Then you find that this TSR doesn't work under that multitsaker unless you load it after running this patch program. With UNIX you just don't get programs stepping on each others' toes. You can occasionally get problems with device drivers, but then the installation tells you you have a conflict, backs out of the installation, and lets you edit your devices file. You can't even do that with MS-DOS, unless the programmer (through the goodness of their heart) condescended to give you a way to fix it... but it's different for *every* program. > >How many times has Fido gotten wedged on you. Oh, that's right. Fido just > >shuts down at 4AM every day so it doesn't have to deal with the problem. > who uses fido? I certainly don't. I don't know of any sites that do. But > since Fido has nasty problems, I guess that means all PC comm apps do too :-) I don't know about other PC communications programs, but Fido is the nearest equivalent in the PC world to UUCP. Repeat after me... UUCP IS NOT NETWORKING. > There are something like 40M machines that can run DOS apps. Yes, and this (to get back to the subject) is responsible for a huge loss of productivity among knowledge workers in this country. You're proud of it? > do they support mice well? Yes. > windows apps? graphical apps? Yes (not on a remote terminal, of course). > TSRs? Probably. I haven't tried any of these... TSRs are one of the things I'm trying to forget about from my days in the DOS world. > `dirty' apps, There's another kind? > dos extended apps? I'm not familiar with this term. Can you provide examples? > cut and paste between apps? Yes. [ i said, the internet virus is a red herring -- the internet is deliberately set up as a low-security network ] > Just because it was designed that way doesn't mean squat. Sure it does. There's no reason you have to set your network up like the internet. That's like saying that, because this guy left his safe unlocked during business hours and got burgled, all safes are useless. > security is only as good as the weakest link. The PC networks do have some > limited security features -- server password protection. Granted that this > isn't a complete solution but the environment is much more limited than say > arpa or uucp. Arpa is a special case... most UNIX networks are not on the internet. Most UNIX networks are local area nets just like most PC networks. And, again, UUCP is not a network. > You are right, you didn't say unusable, just about everything but, though. > You did say that they were isolated environments where data sharing is > done by passing diskettes around -- which I showed via example is not > true at all. Most places it is. You're trying to prove a negative... which you can't do by just giving an example or two. But there were offices doing email and file sharing on UNIX minicomputers as long ago as 1976. DOS equivalents really only got rolling in the last couple of years. I would say that 6-8 years of holding back progress is quite enough. > I grant that Unix is far more elegant in most ways but elegance > isn't everything. You may have disgust for the majority but people do vote > with their $$$. Yes, and they vote for the folks who campaign. Now then, who's got the deepest pockets? The biggest campaign chests? IBM. -- Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation. Biz: peter@ficc.uu.net, +1 713 274 5180. Fun: peter@sugar.hackercorp.com. `-_-' "Seems to me that posting an entire RFC in PostScript is like posting a 'U` Sun-3 binary to comp.sources.unix." -- sgrimm@sun.com (Steven Grimm)
phil@attctc.Dallas.TX.US (Phil Meyer) (10/07/89)
Aside from the flame war brewing here, there are several good points being made. >In article <7971@microsoft.UUCP>, philba@microsoft.UUCP (Phil Barrett) writes: >> There are something like 40M machines that can run DOS apps. > In article <6441@ficc.uu.net> peter@ficc.uu.net (Peter da Silva) writes: >Yes, and this (to get back to the subject) is responsible for a huge loss >of productivity among knowledge workers in this country. You're proud of it? I have been a proponent of 'personal computing' vs. departmental or organizational computing. No more. In 1983 several articles were published concerning the M.I.S. losing control of computing resources. At that time I lost a very large bid to Honeywell because I would not consider that departmental systems with their limited applications could *possibly* have anything to offer to a group of computer literate engineers. I personally lost a possible $100,000.00 in profits. Enough to make me re-think my position. I was bidding a DOD based LAN connected by Novell star LAN servers. This was the top of the line in personal computing environments (and still may be). What did they prefer? A centralized, controlled, and structured environment that supported ONLY the specific tools that they needed to get the job done. It is a myth that DOS has a great wealth of applications. Wealth implies value. Value to whom? If I don't need it, why am I worried about availibility? It has been shown by several studies, (the most recent I have seen is one produced by the American Bar association) that basically there are only four applications in use on Pee Cees. 1. Word Processing 2. Spread Sheats 3. Database Products 4. Desk-top publishing (which I consider an evolution of W.P.) I admit a great number of specialized applications exist, and this is an old topic. I simply wish to demonstrate that 'Personal Computing' may in fact be counter productive. The experiences that I have had would bear this out. The most intellegent systems managers out there are interested in functionality and not personal preference. Give your users the right tools. Don't let them choose their own, and bicker over the merits of each. Most good managers listen to the needs of the whole, and then dictate policy. A network of Pee Cees does not contribute to the standardization, conformance, and productivity of knowledge workers in and of itself. Only carefull planning, organization, and even distribution of the work load can make the system successfull. > >> do they support mice well? > >Yes. > >> windows apps? graphical apps? > >Yes (not on a remote terminal, of course). What about the rise of Xterminals? They will be the future. > >But there were offices doing email and file sharing on UNIX minicomputers >as long ago as 1976. DOS equivalents really only got rolling in the last >couple of years. I would say that 6-8 years of holding back progress is >quite enough. > >> I grant that Unix is far more elegant in most ways but elegance >> isn't everything. You may have disgust for the majority but people do vote >> with their $$$. > >Yes, and they vote for the folks who campaign. Now then, who's got the >deepest pockets? The biggest campaign chests? This is, of course, alt.politics right? The public is only aware of what is shown to it. Color, music, and slogans sell products to the 'general public'. If enough of the right things happen to any product, it can be successfull. Don't kid yourself. The Next box has shown us all this same principal again. What is Job's *real* contribution? The software, of course. What did IBM buy? The software, of course. What do the commercial vendors push? The 'box'. A box that is mediocre at best, but we love it! As Arsineo would say, hmmm... +=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+ | Honesty is in the eye of | Phil Meyer | | the creditor. | Usenet: <backbone>!attctc!lodestar!phil | | | VoiceNet: (214) 991-0897 | +=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+
mpip@uhura.cc.rochester.edu (Robert Annechiarico) (10/14/89)
In article <9605@attctc.Dallas.TX.US> phil@attctc.Dallas.TX.US (Phil Meyer) writes:
- It has been shown by several studies, (the most recent I
- have seen is one produced by the American Bar association) that basically
- there are only four applications in use on Pee Cees.
- 1. Word Processing
- 2. Spread Sheats
- 3. Database Products
- 4. Desk-top publishing (which I consider an evolution of W.P.)
- I admit a great number of specialized applications exist, and this is an old
- topic. I simply wish to demonstrate that 'Personal Computing' may in fact
- be counter productive. The experiences that I have had would bear this out.
- The most intellegent systems managers out there are interested in functionality
- and not personal preference. Give your users the right tools. Don't let
- them choose their own, and bicker over the merits of each. Most good managers
- listen to the needs of the whole, and then dictate policy.
This astonishing, authoritarian attitude is itself a good explanation of
why so many choose personal computers.
Incidentally, you, or the lawyers, left statistical programs off the list.
Only "specialized applications"? So much for omniscience.
Eric Carleen
University of Rochester Medical Center
(716)-275-5391
Bitnet: heartedc@uorhbv
Internet: mpip@uhura.cc.rochester.eduphil@attctc.Dallas.TX.US (Phil Meyer) (10/15/89)
In article <3326@ur-cc.UUCP> mpip@uhura.cc.rochester.edu (Robert Annechiarico) writes: >- Give your users the right tools. Don't let >- them choose their own, and bicker over the merits of each. Most good managers >- listen to the needs of the whole, and then dictate policy. > >This astonishing, authoritarian attitude is itself a good explanation of >why so many choose personal computers. > You forget, or are not considering, that most of the American work force is NOT automated, and have no desire to do so. Talk of 'the computer will destroy the world' is more common than you know. I am talking about automation. Not the fact that a few users at a given site have or want a Pee Cee. Name an organization. What percentage of people at the site use a Pee Cee? I think that everyone will agree that the Firms with total automation (if there really is such a thing) are VERY few! You think too small. Of course you want a Pee Cee. Of course you can benefit from it. But will your company/firm/organization reap the same benefits, or will they even be harmed by you. If you really do well, your peers will dislike you and your tactics. Your superiors will fear your success. This is a real situation. It is only the 'authoritarian' policy of management that can save your career, and help the 'rest' of the workforce come to the same level of proficiency that you possess. An attitude of 'everyone for himself' cannot exist in corporate America, and is the attitude promoted by 'Personal Productivity'. The 'me generation' will fail, unless WE can learn to work with others less enthused about computers and technology than we are. End of soapbox. +=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+ | Honesty is in the eye of | Phil Meyer | | the creditor. | Usenet: <backbone>!attctc!lodestar!phil | | | VoiceNet: (214) 991-0897 | +=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=+
peter@ficc.uu.net (Peter da Silva) (10/15/89)
A case can be made that for a site with a large number of non-computer-
trained people will do much better with a single UNIX box and a number
of dumb terminals. You only need one expert... with PCs everyone has to
be computer-literate.
I don't know about you, but I'm glad my department secretary has no problems
sending me mail when a phone call comes in. Or taking orders in email. Or
posting news articles when someone's out for the day.
--
Peter da Silva, *NIX support guy @ Ferranti International Controls Corporation.
Biz: peter@ficc.uu.net, +1 713 274 5180. Fun: peter@sugar.hackercorp.com. `-_-'
'U`
Quote: Structured Programming is a discipline -- not a straitjacket.mco@hpftc.UUCP (Mark C. Otto) (10/16/89)
In article <9716@attctc.Dallas.TX.US> phil@attctc.Dallas.TX.US (Phil Meyer) writes: >In article <3326@ur-cc.UUCP> mpip@uhura.cc.rochester.edu (Robert Annechiarico) writes: >>- Give your users the right tools. Don't let >>- them choose their own, and bicker over the merits of each. Most good managers >>- listen to the needs of the whole, and then dictate policy. >> >>This astonishing, authoritarian attitude is itself a good explanation of >>why so many choose personal computers. >> > ...[lots of smeary junk deleted]... > >Of course you want a Pee Cee. Of course you can benefit from it. But will >your company/firm/organization reap the same benefits, or will they even >be harmed by you. If you really do well, your peers will dislike you and >your tactics. Your superiors will fear your success. This is a real >situation. It is only the 'authoritarian' policy of management that can save >your career, and help the 'rest' of the workforce come to the same level of >proficiency that you possess. An attitude of 'everyone for himself' cannot >exist in corporate America, and is the attitude promoted by 'Personal >Productivity'. The 'me generation' will fail, unless WE can learn to work >with others less enthused about computers and technology than we are. > >End of soapbox. > Yeah. Right. You sound like the sorta guy that just came back from a thorough indoctrination by Jerry Falwell. You rhetorical questions don't even have question marks on them - as if you were stating a fact. The real fact of the matter is that PEOPLE MUST COUNT INDIVIDUALLY for ANY organization to thrive. Management is simply the task of motivating each individual into WANTING TO CONTRIBUTE their best effort to the task at hand. In a well-managed organization, workers readily accept tools which made their work easier, faster, and more enjoyable - freeing them to explore the higher-level aspects of their job where they can really apply their knowledge and/or skill to the best advantage. Unfortuneately, most office workers are forced to use 'productivity tools' which complicate and debase the very work which they are supposed to simplify and generalize. It is true that the great majority of the semi-automated office workers in business today find themselves working on P.C. products. These P.C.'s are generally purchased for the workers by insecure and generally computer illiterate managers. These managers are convinced that P.C. products are the solution of choice for three reasons - 1. P.C. hardware is cheap, so if their choice is wrong it probably won't damage the company too much (and thus affect their rise to power). 2. P.C. software is plentiful and cheap, so that even if their choice is far from optimal they can probably find a vendor that sells something that they can press into service and then blame the poor software vendor and/or their underlings for poor productivity (and thus not affect their rise to power). 3. The "everyone else is doing it this way so it must be O.K." syndrome. Until just recently "nobody ever got fired for buying XYZ" was the most commonly held belief amongst managers. Only managers who are truely willing to spend the time, money, and personal effort necessary to truely find and aquire adequate tools for their employees (and these people are *VERY* few and far between) will reap the tremendous fruits which can be borne from a well-managed project with frugal staffing and high productivity. These are the mangers that will rise to greatness in well-structured and intelligently run organizations. They quickly earn the respect of their peers, provide tangeable results with a minimum of fuss for their superiors and foster the undying loyalty of their underlings. Managers that are not literate in an area such as computers, yet are forced to make decisions about their purchase, application, viability, etc., are almost doomed from the start. Companies that are smart should provide training before supplying capitol expenditure dollars. I firmly believe that if this had been done prior to companies dumping millions of dollars into P.C.'s, then the world would be a better place for all of us. -- Mark C. Otto Sysop Extrordinare UUCP: {teemc | fmsrl7 | rjf001}!hpftc!mco Voice: 1-313-390-5001 USnail: 15301 Mercantile, Dearborn, MI. 48120 Disclaimer: NOBODY SHOULD EVER BE HELD RESPONSIBLE FOR ANYTHING THEY EVER SAID
hollombe@ttidca.TTI.COM (The Polymath) (10/17/89)
In article <9716@attctc.Dallas.TX.US> phil@attctc.Dallas.TX.US (Phil Meyer) writes: }Of course you want a Pee Cee. Of course you can benefit from it. But will }your company/firm/organization reap the same benefits, or will they even }be harmed by you. If you really do well, your peers will dislike you and }your tactics. Your superiors will fear your success. This is a real }situation. ... I'll give you a real situation. My boss just _gave_ me a PC. I didn't even ask for it. He doesn't fear my success, he needs it. His bonus is directly affected by my productivity. As for my peers, anyone stupid enough to resent me getting ahead by using modern technology won't be my peer for long. The same boss recently gave me a promotion. -- The Polymath (aka: Jerry Hollombe, hollombe@ttidca.tti.com) Illegitimis non Citicorp(+)TTI Carborundum 3100 Ocean Park Blvd. (213) 452-9191, x2483 Santa Monica, CA 90405 {csun|philabs|psivax}!ttidca!hollombe
toma@tekgvs.LABS.TEK.COM (Tom Almy) (10/18/89)
In article <20471@hpftc.UUCP> mco@hpftc.UUCP (Mark C. Otto) writes: >Managers that are not literate in an area such as computers, yet are forced >to make decisions about their purchase, application, viability, etc., are >almost doomed from the start. A *good* manager, when faced with having to make a decision in an area in which he/she is not experienced, will seek the advice of one or more experts and base decisions on that advice. Blind decision making is simply a *bad* management practice. And it's companies with bad management that are doomed from the start. >Companies that are smart should provide training >before supplying capitol expenditure dollars. Yep, another good management decision! Tom Almy toma@tekgvs.labs.tek.com Standard Disclaimers Apply
greg@tcnz2.tcnz.co.nz (super) (10/26/89)
In article <20471@hpftc.UUCP> mco@hpftc.UUCP (Mark C. Otto) writes: [ .. some deleted .. ], most of which I agree with > >It is true that the great majority of the semi-automated office workers in >business today find themselves working on P.C. products. These P.C.'s are >generally purchased for the workers by insecure and generally computer >illiterate managers. These managers are convinced that P.C. products are >the solution of choice for three reasons - > >1. P.C. hardware is cheap, so if their choice is wrong it probably won't > damage the company too much (and thus affect their rise to power). >2. P.C. software is plentiful and cheap, so that even if their choice is > far from optimal they can probably find a vendor that sells something > that they can press into service and then blame the poor software vendor > and/or their underlings for poor productivity (and thus not affect their > rise to power). >3. The "everyone else is doing it this way so it must be O.K." syndrome. > Until just recently "nobody ever got fired for buying XYZ" was the most > commonly held belief amongst managers. > This is a fairly major generalization. We have rather enlighted management, but we got shafted from the UK Head Office. Out with networked Unix solution, replace with ps/2's. Why - 1) Worldwide buying strategies count for something. In the UK market, lanned pc's are where it is at, and the global manager has been bit by Unix before. He is computer literate, etc, but has a global situation to address. 2) the bulk of users need spreadsheets and wp, and dos packages are more robust in the hands of the naive user, especially when in a small branch. 3) The airlines have some important global reservation systems. They run on ps2's under dos or os2. We have to run it. I wanted to try it under xenix but no-one was interested. 4) The age old IT managers argument - in 2000, who will be around ? IBM will. 5) Unix bad press. OSF really isn't helping the big MIS managers perception of Unix as a serious product. There are still too many individual unixes without ABI's or ANDF or whatever else to make Open Systems a believable concept. They seem more prepared to listen to IBM's promises of a coherent os2 future than 20 other vendors talking about how unix will get it's shit together real soon now. In some ways, I can't blame them. Some of this makes very good business sense, much as I hate the whole idea. Time to find a more appropriate job. I love Unix, os2 seems to be a toy. My question is - does dos/os2 have enough of the important and useful unix tools such as make, sccs, grep, awk, sh, etc, for people to work productively under ? In the UK, a workstation is a PC. I want a workstation, but a Sun or 3100 or AViiON seem to be alien to these peoples thinking. I was offered all the processing power I would ever need, and they were talking 386's. No smiley, even. Disclaimer - These depressing thoughts are mine alone. Think I'll go suicide now. -- Greg Calkin Thomas Cook N.Z. Limited, ...!uunet!vuwcomp!dsiramd!marcamd!tcnz2!greg PO Box 24, Auckland CPO, or greg@tcnz.co.nz New Zealand. Phone (09)-793920
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/27/89)
In article <164@tcnz2.tcnz.co.nz>, greg@tcnz2.tcnz.co.nz (super) writes: | 4) The age old IT managers argument - in 2000, who will be around ? IBM will. And AT&T will vanish? OSF I can believe will go away, it's a conglomerate of natural competitors, but somehow I can't picture AT&T as anything but a natural force. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
deem@interlan.UUCP (Mike Deem) (10/28/89)
In article <164@tcnz2.tcnz.co.nz> greg@tcnz2.UUCP (super) writes: > >My question is - does dos/os2 have enough of the important and useful unix >tools such as make, sccs, grep, awk, sh, etc, for people to work productively >under ? > DOS and UNIX don't compare, however I will answer this question concerning OS/2 and UNIX. Your question could be asked in two parts: "Is OS/2 pro- ductive for users?" and "Is OS/2 productive for developers?" because we (the developers) really use computers very differently then them (user). For the user, OS/2 is a preferable environment over UNIX for a rather simple reason: it was designed to be easy to use. UNIX on the otherhand was designed to be easy to develop under. As a developer, you find it easy to get things done under UNIX, however do your (novice) users find it as easy to use your UNIX based systems? As a developer, I find OS/2 to be as good as UNIX. I must admit though that I havn't done much work in UNIX. I moved from DOS to OS/2. There are tools available that make OS/2 _VERY_ UNIX like. True, they don't come with the OS/2, but why should users pay for a bunch of stuff used only by programmers. I use the C Shell for OS/2 from Hamiliton Laboratories along with a number of other utilites I have picked up here and there and find that I can get as much programming, debugging and testing done on OS/2 as anyone on UNIX can. Because OS/2 is a new operating system, it still lacks some things the UNIX world takes for granted. One major area is wide area networking. I still log into a UNIX machine to read my news for example. Local area networking on the other hand is comming along nicly. I use CC:Mail on OS/2 to send and receive mail from DOS, UNIX and MAC users. Mike Deem Racal InterLan deem@interlan.interlan.com sun!interlan!deem The above opinions are my own and do not necessarly reflect those of Racal InterLan.
aglew@urbana.mcd.mot.com (Andy-Krazy-Glew) (10/29/89)
>A case can be made that for a site with a large number of non-computer- >trained people will do much better with a single UNIX box and a number >of dumb terminals. You only need one expert... with PCs everyone has to >be computer-literate. Dumb terminals hell! How about a single UNIX box with a number of bitmapped graphics terminals? Give 'em a MAC interface... I've been trying to get this for a long time. Back when I was at Gould, working on a machine with enough CPU and I/O to conceivably handle 100s of bit-mapped screens at a time, if only there where a way of displaying the bits, I wrote memos about it. Well, finally it's been done. The SUNRIVER for *86 PCs, X terminals, and Motorola's own multiple tubes per box make it possible. -- Andy "Krazy" Glew, Motorola MCD, aglew@urbana.mcd.mot.com 1101 E. University, Urbana, IL 61801, USA. {uunet!,}uiucuxc!udc!aglew My opinions are my own; I indicate my company only so that the reader may account for any possible bias I may have towards our products.