koren@hpfelg.HP.COM (Steve Koren) (12/01/89)
Since there are a few questions which are asked quite frequently about
SKsh, I thought I might post some answers here (this is mainly to give
me some hope of catching up on my email before it fills up the remaining
disk space on my system :-). Also, I have around a dozen replies to
questions like these which bounced.
1) FTPing SKsh
Due to security restrictions here, I cannot make SKsh available for
FTP. Sorry. I think that some comp.binaries.amiga archive sites
will have it, although I do not know where these are.
2) emailing SKsh
I'd like to avoid this due to the size of the uuencoded zoo file. I
may agree to it in a few unusual cases, but I am much more likely to
send it if you send me A) a blank 3 1/2 " diskette, and B) a
self-addressed stamped mailer of some sort. To appriciate why I
don't like to email it around, consider that since I first posted
SKsh (which was Ash back then) I have gotten over 50 requests to
email it. At roughly 1/4 megabyte for the zoo'ed and uumangled file,
this is alot of bits. Perhaps to alleviate this problem and make
SKsh easier to get, some kind sole could post a reply to this note
which names a machine where the latest SKsh is likely to be, and can
be FTP'ed from.
3) Obtaining source code
This is probably 2nd or 3rd on the list of most frequently asked
questions. The short answer is that, yes, I will eventually be
releasing source code. The long answer is that I am waiting, for the
reasons described below:
a) For the moment, I desparately wish to avoid branches in the
version stream. At the moment, if someone tells me that they are
using version 1.2, I know *exactly* which system they have, and I
can say intelligent things like "the bug you report is known about
and was fixed in the next release". Otherwise, someone will have
SKsh modified by John, extended by Bert and ported to the Timex
Sinclair by Frank, and I will have no clue how to help these
people when they have problems. I learned from experience (with
QRT) that I end up spending alot of time dealing with versions I
didn't write once I release source code.
b) The actual process of obtaining the source code is somewhat
compex; you cannot just plunk down some .c and .h files and
compile. First of all, you need to have a working yacc. Then,
you need to have some reusable code libraries I have developed to
implement things like hash tables. Finally, you need to install
the whole set of test scripts, many of which have dependencies to
my specific file structure. I have to hunt down these
dependencies and eliminate or document them before I can release
the source.
c) There are many parts of the source that I have yet to comment
adequately.
d) I need to write up a set of guidelines for what you need to do to
release a modified SKsh. For example, I will probably *require*
that the entire set of test scripts be run with *no errors* on the
modified version. This is to attempt to insure some moderate
level of robustness; ie, that your change doesn't break something
that you're not aware of. Also, I will definately come up with a
minimum set of required files. Again from QRT I have found that
around 60% of all requests are of the form, "I have an incomplete
system, could you send me a complete one?" I have never minded
doing this, but I'd like to reduce the frequency of those requests
as far as possible, especially because many of them end up being
US-mailed at my expense for disks and postage. Hence, the required
list of files.
So, source will be coming, but not very soon. I will probably wait
until Sksh is, from my point of view, fishished. That is, until I
have added everything that I realisticlly hoped to add. I have no
idea when this will be. I may make exceptions and give the source to
a *few* people who have offered to help debug/enhance/etc, but I will
ask these people to not release modified versions until the source
code has been distributed publicly.
I hope this does not seem unreasonable; I'm only trying to be as
efficient as possible in creating a high quality piece of software.
I have to do certain things to reduce the load on my time, since I
have only a fraction of my spare time to dedicate to SKsh (I like to
do other things, too :-)
Again I'd like to thank everyone who has sent mail with suggestions,
bug reports, etc. If you didn't get a reply, it bounced.
- stevetadguy@cs.odu.edu (Tad Guy) (12/08/89)
In article <13920022@hpfelg.HP.COM> koren@hpfelg.HP.COM (Steve Koren) writes: > Due to security restrictions here, I cannot make SKsh available for > FTP. Sorry. I think that some comp.binaries.amiga archive sites > will have it, although I do not know where these are. xanth.cs.odu.edu in /usenet/comp.binaries.amiga/volume89/unix: -r--r--r-- 1 news news 57913 Nov 12 23:46 ash.uu1.Z -r--r--r-- 1 news news 57563 Nov 12 23:49 ash.uu2.Z -r--r--r-- 1 news news 56289 Nov 12 23:51 ash.uu3.Z -r--r--r-- 1 news news 30039 Nov 12 23:53 ash.uu4.Z ...tad
koren@hpfelg.HP.COM (Steve Koren) (12/11/89)
> > Due to security restrictions here, I cannot make SKsh available for >> FTP. Sorry. I think that some comp.binaries.amiga archive sites > > will have it, although I do not know where these are. xan> th.cs.odu.edu in /usenet/comp.binaries.amiga/volume89/unix: -r> --r--r-- 1 news news 57913 Nov 12 23:46 ash.uu1.Z -r-> -r--r-- 1 news news 57563 Nov 12 23:49 ash.uu2.Z -r--> r--r-- 1 news news 56289 Nov 12 23:51 ash.uu3.Z -r--r> --r-- 1 news news 30039 Nov 12 23:53 ash.uu4.Z Thanks for the info. That looks like an old (1.0 or 1.1) version. I'd like to actively avoid having people use the old versions, since they contained a number of bugs that were fixed in 1.2. Does this archive have the newer (SKsh1.2) version, also? - steve
pawn@wpi.wpi.edu (Kevin Goroway) (12/15/89)
In article <TADGUY.89Dec7154957@diatomite.cs.odu.edu> tadguy@cs.odu.edu (Tad Guy) writes: >xanth.cs.odu.edu in /usenet/comp.binaries.amiga/volume89/unix: > > -r--r--r-- 1 news news 57913 Nov 12 23:46 ash.uu1.Z [etc...] > ...tad I believe that these are indeed ash, and not Sksh. Am I correct? -- +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+= | Worcester Polytechnic Institute | "It happens sometimes, people just | | Pawn@wpi.wpi.edu Pawn@wpi.bitnet | explode, natural causes."-Repo Man | +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
tadguy@cs.odu.edu (Tad Guy) (12/18/89)
In article <6307@wpi.wpi.edu> pawn@wpi.wpi.edu (Kevin Goroway) writes: > In article <TADGUY.89Dec7154957@diatomite.cs.odu.edu> tadguy@cs.odu.edu (Tad Guy) writes: > >xanth.cs.odu.edu in /usenet/comp.binaries.amiga/volume89/unix: > > -r--r--r-- 1 news news 57913 Nov 12 23:46 ash.uu1.Z > > I believe that these are indeed ash, and not Sksh. Am I correct? Perhaps. However, the time/date look right, and the author has the same name. Regardless, this one is obsolete. The most current version available on xanth is now /amiga/SKsh-1.2.zoo ... ...tad