[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Chicken-hackers] let the Windows binary go bye-bye

From: Brandon J. Van Every
Subject: Re: [Chicken-hackers] let the Windows binary go bye-bye
Date: Tue, 07 Nov 2006 23:21:40 -0800
User-agent: Thunderbird (Windows/20060909)

John Cowan wrote:
Brandon J. Van Every scripsit:

I don't see why you conclude this.  People on chicken-users have
subscribed to the mailing list, which they probably accessed from the
Chicken homepage.  People like myself routinely do due diligence and
ask questions before they get themselves into further messes.

If only more developers were like you!  Most seem to just plow ahead
until they run into the first roadblock, then give up.  And it's not
just developers -- googling for "read the instructions|directions first"
produces more than 30,000 hits.

If developers *aren't* like me, to some extent, they're never going to find Chicken in the 1st place.   Know Thy Demographic.

The 1st direction in INSTALL-CMake.txt is "If in trouble, please contact
me through the Chicken mailing list.  You can subscribe to the list
from the Chicken homepage,"
I noticed there was no such directives in README or README.darcs,
so I just added them.

Did you ever take that test in elementary school that begins "Before
doing anything, read all the directions carefully"?

Yep.  But if you don't pay attention to READMEs, it's your own damn fault.  The kind of person who can successfully use Chicken, is the kind of person who knows how to DIY in open source land.  We're not out to create clickable installers for MS Corp types yet.  Chicken isn't that advanced in the plug 'n' play dept. yet.  It's pretty advanced for open source DIY types though.  It builds cleanly, the process is documented, it works, it's simple, etc.

There's no point asking people who like their C++ / Java / C# /
.NET just fine, thank you very much!  They'll just say, "Why is your
language so brain dead that you're even asking this?"  And they'd have
a point.  The answer is, "Because Chicken is a small project with few
people working on it."  And then they'll say, "So if you're so small,
why should we care?"  And they'd have a point.

No argument there; however, specific familiarity with Scheme is *not*
a requirement.  Look to people who use Perl/Python/Ruby/Lua for Windows,
say, and ask them.

But we already know the answer.  Chicken is supposed to have installable binaries to any directory.  We don't have it and we're not implementing it.  "In what manner would you like your Chicken broken?" just really isn't a question worth asking those crowds.  Their expectations are based on thousands of project contributors; Chicken is not at that scale.

In contrast, people with experience with boutique High Level Languages
are more likely to understand the research / implementation / support
difficulties of keeping such projects alive.  Of course, those people
can read compilation instructions, and it all works as advertized,
so why are we asking them?

The question is not whether they can, but whether they will bother.
And that calls for an empirically tested answer.

Go test it then.  It's a waste of time.  I've already weighed in with my opinion: an excellent build process trumps a broken binary.

Chicken isn't ready for mass marketing or mass acceptance.  That's just

No Scheme implementation ever will be.  That's reality too.

I don't agree.  PHP became famous because it did the Web.  What's required for a Scheme to be popular, is for a specific Scheme to do 1 particular application area really really well.  Common Lisp has suffered in the mindshare wars because it doesn't do any particular thing particularly well.  It's general purpose and the CL community has the collective blinder that it's supposed to be general purpose.  A specific Scheme implementation is culturally / politically far more capable of "going its own way."

For instance, Bigloo is performance oriented and does C, Java, and C#.  It could conceivably carve out a .NET niche if it wanted to.  Kawa Scheme is Java oriented, it could conceivably provide a better embedded language for Java if it wanted to.  Or ace some application domain that needs that as a prerequisite.  It remains to be seen what Chicken will aspire to do well.  I have my own ideas about game development, but who knows what I'll actually get done.

Brandon Van Every

reply via email to

[Prev in Thread] Current Thread [Next in Thread]