jimr@maths.su.oz.au (Jim Richardson) (07/23/90)
In article <5570451@hpfcdc.HP.COM> in comp.sys.hp, rer@hpfcdc.HP.COM (Rob Robason) writes: >These comments are absolutely my own and in no way reflect HP's >position: > >I think that the need to provide some sort of high speed link between >our support customers and our support organization is self evident. HP >has a long standing tradition of good support, and we are consistently >ranked in the highest eshelons in this regard. But we are not so naive >as to think that we need not improve to keep our leadership position, >nor that we do not have problems to iron out. As an Apollo customer, I feel that great improvements are urgently needed in many aspects of support. >... >I think a couple of non-trivial stumbling blocks stand in the way of >HP's use of the net for support purposes: > >1) The commercial (i.e. not for gain) use of the internet is > apparently forbidden. Without wishing to cast dispersion on any of > our competition, HP is not a company that will knowingly violate > this policy. Work to change, yes, but not violate. I addressed this in my posting <1990Jul18.102038.22220@metro.ucc.su.OZ.AU>, where I argued that the NSFNET acceptable use guidelines do *not* prohibit the kind of activities we are asking HP to undertake on the network. In particular, I said: ]In any case, the statement invites review by the NSFNET Project Office. Has ]HP made a submission to the Office for review? If so, could we be told what ]the Office said? Even if HP *has* asked in the past and failed, I suggest that ]a well structured *JOINT* submission from HP *and* interested US academic ]institutions could well be successful. I'd still like to hear what HP has done so far to "work to change", in Rob Robason's words. I'd also like to encourage people at US research or academic institutions, which have standing with NSFNET, to think about mounting a case to the Project Office. Those connected to other networks might think what steps they could take: I've made some enquiries myself about the Australian AARNET network which are quite encouraging. >... > While small companies may get away with basing their whole support > system on the net, when a company the size of HP steps in and does > the same, they draw a lot more attention and are more likely to > bring cries of foul-play from competitors, gateways and net purists. Without disrespect for the size of HP, I'd like to point out that Sun and Apple are not tiddlers either. Sun has an archive on uunet.uu.net. And as Jim Wright has pointed out, mjohnson@Apple.COM (Mark B. Johnson) recently announced in comp.sys.mac.misc that Apple will soon be making a wide range of system development software available on a new FTP archive. What we're asking for in an HP archive in Request 3 of the Open Letter to HP is considerably more modest! Dave Taylor has also noted in <1002@limbo.Intuitive.Com> that ) - Bug reports and defect reports can be sent via email to ) at least: Sun, DEC, MIPS and Apple... Does anyone know of other examples? >2) It doesn't take a genius or an insider to figure out that about 1/5 > to 1/4 of HP's computer revenues are from support contracts. For a > $12 billion company, that's not something you treat lightly. I'm > sure the folks in support are wrestling with how to provide these > services on a public network to a restricted set of potential users > (i.e. those who've purchased support contracts). It wouldn't make > sense to give away to some what you're charging others $2 Billion > for, you wouldn't sell many contracts on those terms. It should be possible to support paying customers by email on a public network, provided those customers are using the equipment being supported for purposes acceptable to the network. As far as a public archive is concerned, HP would be giving something away for free, I agree. But in the long term this could be beneficial to HP's profits rather than the opposite. Firstly, making information available by FTP to everyone on the net is much cheaper than distributing it by old-fashioned means. Secondly, if routine support questions, which hundreds of different customers might want to ask, can be answered by the customers themselves through reference to a public archive, HP support resources will be freed to process tricky non-routine questions more effectively. That should make life more interesting for support engineers, apart from anything else :-) Thirdly, an efficient and attractive support system available via Internet will be a big plus for HP in selling more machines. >3) There is an incredibly intricate support organization in place to > support HP's miriad products from countless divisions in multiple > business sectors. A major accomplishment has been the ability to > provide this support in a mostly consistent way to customers to hide > the fragmentation behind the scenes. ... "Hide"? Many Apollo customers at least are quite aware of this "fragmentation"! > ...A change such as support via a > public network would require significant coordination between lots > of HP entities, many of whom have very different goals. Such > coordination isn't easy even in small groups, and takes time to meet > everyones needs and fit everyones schedule. Sure, it's going to take a lot of work for HP. But if HP can't move rapidly to take advantage of the networking technology that Dave Taylor has called "the wave of the future", then there are other companies who will. >... >As an aside, I wouldn't hold my breath for any free support though, I >don't think that's in the future. Any network patch system you see, >you'll only see if you pay for it. See above. We pay for support contracts to get major software upgrades, and certainly expect to continue to do so. It may require some changes in corporate culture, but I don't see why a responsible company, proud of its reputation for fine products -- as HP justifiably is -- should not be willing to make incremental patches publicly available when there do turn out to be blemishes in its software. >Rob "My $2 Billion worth" Robason -- Jim Richardson Department of Pure Mathematics, University of Sydney, NSW 2006, Australia Internet: jimr@maths.su.oz.au ACSNET: jimr@maths.su.oz FAX: +61 2 692 4534 -- Open Letter: The final version has been delayed a little, but should be ready by early next week if not sooner. There is a volunteer to collect signatures in Europe (thanks, Colin!); another for North America would be much appreciated.
bkgray@encore.kent.edu (Brian Gray) (07/23/90)
In article <1990Jul23.100007.2429@metro.ucc.su.OZ.AU> jimr@maths.su.oz.au (Jim Richardson) writes: > >It should be possible to support paying customers by email on a public >network, provided those customers are using the equipment being supported >for purposes acceptable to the network. > But what about customers who don't meet the "acceptable purpose" test? It seems to me that you are asking HP to develop and maintain one support method for academic/research installations and another for comercial (for-profit) installations. If I worked for HP and were involved in providing support services, I would want methods that works for ALL customers. It seems that customer access to the HP internal network via a local HP sales office machine would come closer to serving the needs of both for-profit and non-profit installations. >Secondly, if routine support questions, which hundreds of different customers >might want to ask, can be answered by the customers themselves through >reference to a public archive, HP support resources will be freed to process >tricky non-routine questions more effectively. That should make life more >interesting for support engineers, apart from anything else :-) > That's what SupportLine does (although HP needs to keep it a bit more up-to-date). brian
jwright@cfht.hawaii.edu (Jim Wright) (07/24/90)
jimr@maths.su.oz.au (Jim Richardson) writes: >Without disrespect for the size of HP, I'd like to point out that Sun and >Apple are not tiddlers either. Sun has an archive on uunet.uu.net. And as >Jim Wright has pointed out, mjohnson@Apple.COM (Mark B. Johnson) recently >announced in comp.sys.mac.misc that Apple will soon be making a wide range >of system development software available on a new FTP archive. Minor nit: It is not a new site. Apple has offered this service for quite a while. But it has not been updated since Jan or Feb 90. The posting was an announcement that Apple will soon be updating things. -- Jim Wright jwright@quonset.cfht.hawaii.edu Canada-France-Hawaii Telescope Corp.
wunder@hp-ses.SDE.HP.COM (Walter Underwood) (07/25/90)
I addressed this in my posting <1990Jul18.102038.22220@metro.ucc.su.OZ.AU>, where I argued that the NSFNET acceptable use guidelines do *not* prohibit the kind of activities we are asking HP to undertake on the network. In particular, I said: ]In any case, the statement invites review by the NSFNET Project ]Office. Has HP made a submission to the Office for review? If ]so, could we be told what the Office said? Even if HP *has* asked ]in the past and failed, I suggest that a well structured *JOINT* ]submission from HP *and* interested US academic institutions could ]well be successful. I agree that a proposal to support only academic institutions, only in the US, would probably fly. It might not even need review, since it is pretty clearly within the limits of the policy. I don't think that is adequate. We have a *lot* of customers that need this support, and only a small percentage of them are at research institutions in the US. We must address reasearch and commercial institutions world-wide -- that is our customer base. Right now, we have private IP links to some customers, but that is obviously not a long-term solution. I *believe* in IP networks. I ran the project to build HP's IP network (now at 20K hosts). I also think that the use restrictions on the charity networks are a royal pain in the ass. They are reasonable, because the US government has decided that it wants those tax dollars to support research in the US. As a result, CSNET and NFSNET are very useful the people in HP doing research, and almost useless to the people trying to support customers. That is exactly what the National Science Foundation intends. Since the Commerce Department doesn't seem interested in building a subsidised network, it is up to private companies to build IP networks for commercial use. HP is checking out the commercial IP networks (ALTERNET and PSI) so that we can do business over IP. wunder
jimr@maths.su.oz.au (Jim Richardson) (07/30/90)
In article <1990Jul23.120147.25435@math-cs.kent.edu>, bkgray@encore.kent.edu (Brian Gray) writes: >But what about customers who don't meet the "acceptable purpose" test? >It seems to me that you are asking HP to develop and maintain one support >method for academic/research installations and another for comercial >(for-profit) installations. If I worked for HP and were involved in >providing support services, I would want methods that works for ALL >customers. ... Rob Robason has told us that HP is a $12 billion company. I doubt it is beyond them to provide a variety of different support mechanisms to suit their very diverse customer base. > ... It seems that customer access to the HP internal network >via a local HP sales office machine would come closer to serving the >needs of both for-profit and non-profit installations. > >>Secondly, if routine support questions, which hundreds of different customers >>might want to ask, can be answered by the customers themselves through >>reference to a public archive, HP support resources will be freed to process >>tricky non-routine questions more effectively. ... > >That's what SupportLine does (although HP needs to keep it a bit more >up-to-date). I haven't been able to find out much about SupportLine: it seems to allow dial-up access to some sort of database in your local HP office. My sales rep here in Australia is trying to arrange something, but the service does not seem to have a high profile here. There have been a few references to it in the newsgroups before, but no one has sounded very enthusiastic: note Brian's own reservation. Is this service available and relevant to Apollo customers? I'd be interested to see a description of SupportLine from someone who uses it. A dial-up service, say via UUCP, to a local HP office machine could be helpful, if some provisos were met, e.g. materials up-to-date at all offices; enough phone lines; sufficiently high data-transfer rate. I suppose I would use such a service if nothing better were available, but for me it would come as a poor second in convenience, flexibility and maintainability to an Internet email service and FTP archive. HP should be encouraged to adopt a diversity of approaches: Internet support for those customers with access and rights to use it; dial-up or newer alternative networking approaches for other customers. Walter Underwood (wunder@hp-ses.SDE.HP.COM) gives us cause to hope that HP *is* looking at a variety of methods in article <920065@hp-ses.SDE.HP.COM>, where he writes: >I agree that a proposal to support only academic institutions, only in >the US, would probably fly. It might not even need review, since it >is pretty clearly within the limits of the policy. [This could work in other countries as well, either via public international links, or through HP affiliating with and linking to the relevant national networks, such as AARNET in Australia.] >I don't think that is adequate. We have a *lot* of customers that >need this support, and only a small percentage of them are at research >institutions in the US. We must address reasearch and commercial >institutions world-wide -- that is our customer base. Right now, we >have private IP links to some customers, but that is obviously not a >long-term solution. >... >Since the Commerce Department doesn't seem interested in building a >subsidised network, it is up to private companies to build IP networks >for commercial use. HP is checking out the commercial IP networks >(ALTERNET and PSI) so that we can do business over IP. This is good, but it will take time, and many customers, especially outside the US, who can use the Internet will not have access to such commercial networks. So you still won't cover everybody. HP should be moving quickly, to catch up with Sun, Apple, and other companies who are already using the Internet for support. A public FTP archive, main- tained by HP and freely accessible from the Internet, could be set up now, without breaching commercialism guidelines, and would be a big step in the right direction. Open Letter: The final version should be out by Wednesday. Again, apologies for the delay: start of semester here has kept me busy. -- Jim Richardson Department of Pure Mathematics, University of Sydney, NSW 2006, Australia Internet: jimr@maths.su.oz.au ACSNET: jimr@maths.su.oz FAX: +61 2 692 4534
jimr@maths.su.oz.au (Jim Richardson) (08/07/90)
In article <1480002@hparc0.HP.COM> in comp.sys.apollo, graham@hparc0.HP.COM (Graham Eddy) writes (replying to an earlier article of mine): > ... in fact, it would be helpful if people with constructive >criticism contacted *us* rather than the media first. As Mike Gschwind and Len DiPaoloa have already pointed out, a lot of us *have* contacted HP with complaints first before going public. Part of the reason we are upset is that private complaints do not seem to have been heard. See below for my own case. > the merger of the >support structures has not been as smooth as it might have been :-( and >some feedback from you people needing support would help set priorities. The HP/Apollo merger took place more than a year ago. The merger of support structures in the Australian Response Centre took place last December. There has been a lot of time for the problems to be ironed out. In my opinion, this excuse is wearing very thin. >... >> It should be possible to support paying customers by email ... >it is only a minority of customers who have access to the sort of >facilities being touted here. this suggestion thus is based on the >assumption that large and rich customers should receive better service >than smaller customers. ecch! Educational customers, who form the majority of those with access to the Internet, are often *not* as rich as large commercial customers when it comes to finding funds for support. See the comments in the Open Letter on this. So I'm glad to hear that poor customers should get good service too! :-) >> Firstly, making information available by FTP to everyone on the net is >> much cheaper than distributing it by old-fashioned means. > >see above. the suggestion advocates two separate delivery mechanisms. >anyone with a bit of business acumen will realise that this means it >is actually more expensive, not cheaper, if the labour component is >nontrivial. labour is the most expensive component today! Agreed, setting up an Internet Liaison Unit will cost HP money for staff and perhaps equipment. But just the savings on individual SEs having to copy patch tapes themselves (as apparently happens at the Australian Response Centre for one) would go some way towards freeing up personnel to run an FTP archive. Anyhow, remember that HP is a $12 billion company. There are plenty of resources somewhere. If you need to cut something, how about that HP/Apollo 9000 roadshow with flashing lights and "talking" workstations? >... >has anyone making this proposal actually costed an appropriately sized >anonymous ftp server, or is it assumed costless? Workstations are not expensive these days, you know! Anyhow, hpcvaaz.cv.hp.com is already acting as a public FTP server (see <101950138@hpcvlx.cv.hp.com> in comp.sys.hp). You might need an extra disk ... you'll find that third-party ones are quite cheap. :-) >hp/apollo is offering this service already through SupportLine. from >some of the stuff i've seen on the net, it needs improvement. fine, so >let hp/apollo know what needs changing! I asked about SupportLine in <1990Jul30.072518.8055@metro.ucc.su.OZ.AU> -- that article seems to have crossed with Graham's. Could you please give us more details? In particular, what does SupportLine offer for Apollo customers? I've just received details of various HP ApolloLine service options, and SupportLine is not mentioned, at least not by that name. Perhaps it could be what some of us are looking for, especially people without direct Internet access. >> See above. We pay for support contracts to get major software upgrades, >> and certainly expect to continue to do so. It may require some changes >> in corporate culture, but I don't see why a responsible company, proud >> of its reputation for fine products -- as HP justifiably is -- should >> not be willing to make incremental patches publicly available when there >> do turn out to be blemishes in its software. > >*sigh* is it only development people who understand the 80/20 rule? >there are two choices: release software when it is nearly done (i.e. >possibly some obscure bugs) and get it out now, or spend an extra two >(maybe three) years and get it just right. it is better for *everyone* >that reasonable care be taken, rather than perfectionist care, else >it will never be seen at all! in fact, i continually argue that we >should put out software earlier in its cycle, labelled with a clear >warning that it is indeed early and prone to errors - but consumer >gets to decide whether to use it. As far as I know, nobody has accused HP of perfectionism! Agreed, software often has to be released when it's only (say) 80% ready. But at present I don't feel we're seeing enough action on the remaining 20%, which should be covered by prompt attention to APR bug reports, and by properly distributed patches where necessary. >okay, in summary: it is clear that some people are not receiving the >level of support they expect. i am amazed to see people complaining >that hp/apollo does not read notes/usenet, yet these same people post >their complaints there instead of letting hp/apollo know! please, if >you have constructive criticisms, contact your local response centre >or support office: how else do they know to lift their game? *I* am amazed to see such a statement from someone working in the same Response Centre as the support engineer who handles my APRs and to whom I have been complaining privately for months about lack of answers and even acknowledgements. Before the Response Centre took over responsibility for Apollo Product Reports (on 11 December 1989), I had already sent a letter expressing dissatisfaction with the APR service to my sales rep (26 September 1989), followed by phone calls, and a FAX complaining specifically about lack of email response from apr_cs_admin@apollo.hp.com (1 November). The first reply I got said that the level of support contract we had (SSS) did not entitle us to acknowledgement of APRs. After I pointed out that this was incorrect, I received another letter accepting this and telling me the new procedure for submitting APRs to the Response Centre. I started doing this in late December. I FAXed the Response Centre on 2 February and 26 March pointing out that APRs had not been acknowledged, and sent email complaining about lack of acknow- ledgement and response to the SE on 14 May, 18 June, 11 July and 23 July. My first "netpower" posting appeared on 6 July. The reasons for my complaints were as follows. Of the 16 APRs (excluding duplicates) I have submitted to the Response Centre, I have received a formal written acknowledgement by post to only ONE; three others were closed over the telephone. Apart from those, I had not been informed of the official Product Report numbers for any until 13 July, when I was given one official number by phone; four other numbers were emailed to me on 20 July, together with responses (two incomplete) generated in Chelmsford as far back as May. There are still three APRs open, submitted in February and March, for which I have not yet been given official PR numbers. The Australian Response Centre has been sending the APRs to Chelmsford, and they have received and worked on at least some of them. One major problem seems to be that the Apollo International Coordinator and/or the APR admin people in Chelsmford have not been sending acknowledgements or responses back either to me or the SE. The responses that I have received come from the SE who has been extracting them direct from the on-line database (he seems to have started doing this in early July -- that's very welcome, if belated). There's some good news. On 31 July I submitted a new APR (well, actually it was an old one I had sent in August 1989 about 10.1, but that was never acknowledged and the bug's still there in 10.2, so why not? :-). The official PR number reached me via Melbourne from Chelmsford the next day, 1 August. So complaining helps! But was it the months of private complaint or the weeks of public campaigning that did it? Sorry for the tirade. But the point is ... I (and I don't think I'm alone) complained direct to HP till I was blue in the face. Lack of improvement caused the frustration that has led many of us to campaign on the net. -- Jim Richardson Department of Pure Mathematics, University of Sydney, NSW 2006, Australia Internet: jimr@maths.su.oz.au Phone: +61 2 692 2232 FAX: +61 2 692 4534 -- Please keep your OPEN LETTER signatures coming in! Progress report soon.