> I think we need to nail the intro / setup experience and I'm nailing > my colors to Leiningen. I think that needs to be adopted as the > default, standard way to get up and running on Clojure and all the > official tutorials need to be updated to reflect that.
I think getting an experienced Clojurian to agree with that is worth any (hopefully fixed) logorrhea! If people can agree to that one point, I'm really not going to worry about \any of the others (hence I put it first). I think it's an important time to be making it (so let's forget about the other stuff for now). Clojure was \just put on Heroku----and for the first time in a while, it showed up under Google news. Steve Yegge is also (threatening? promising?) to blog about it (after writing an intro to the Joy of Clojure), and he happens to be one of the 2 or 3 most-read language bloggers. That means that a lot of attention is going to be coming Clojure's way soon. It would be very much a travesty if large portions of these people were put off by out- of-date stuff on the website. Once they are put off once, it is going to be much harder to get them in a second time. Speaking to people who I think would not mind being paid to program Clojure----we all have a stake in making the language popular and useful. (As for Steve Yegge----is he reading all this?----if he's totally wrong, then of course people should feel free to disagree with him, and forget about the consequences. But if he happens to be \right, and I do think he mostly is, then making basically dismissive "firm stands" against him is not going to do anybody any good. This isn't a political party, thank God.) It may be that I am really talking about the website (clojure.org, not any of the auxiliary ones, which are a bit of a mess in themselves) more than the language itself. If people receive the \right instructions, setting up Emacs/Leiningen/Web servers etc. is actually not so hard. The trouble is that all of this information is currently scattered to the four winds (I include things like the Paredit cheat sheet, Slime commands, which Emacs to use, etc.), and I don't think we should rely on users to pull this information together themselves---- and at any rate, why should they? Mainstream languages (Java, Python) provide good resources drawn together onto their websites, painless documentation systems for contributed packages (where is the Contrib documentation these days?), and they have a pretty standard option for getting started. I think Clojure should fully aspire to that, and provide ways for users to help with developing it. That is a long-term project, but in the meantime, there seem to be some really simple things that can be done now. That's not my whole point by any means, but it's more than enough. I wish that someone of the stature of Phil Hagelberg would have brought it up on this thread; but as I have made (enough of) my point, I'll get off my hobby-horse now. See (some of) you at the Bay Area meetup tomorrow! Nick. On Jul 6, 11:37 pm, Sean Corfield <seancorfi...@gmail.com> wrote: > Much better. Now I can read it and see your points... and respond... > > On Wed, Jul 6, 2011 at 10:42 PM, nchubrich <nchubr...@gmail.com> wrote: > > * Clojure still ends up turning off new users more than it needs to. > > I think we need to nail the intro / setup experience and I'm nailing > my colors to Leiningen. I think that needs to be adopted as the > default, standard way to get up and running on Clojure and all the > official tutorials need to be updated to reflect that. It's the > biggest, single roadblock IMO and all that nonsense about downloading > ZIP files and running some specific Java command is a huge barrier to > entry. So far everyone I've introduced to Clojure has struggled with > the Java infrastructure stuff but has "got" Leiningen instantly. > > > * It also can do a better job of attracting and retaining core > > contributors. I cited an example of someone who posted a patch to > > make refs persistent. She ended up being ignored, and left for > > Erlang. But Clojure needs people like her. > > Unfortunate but all open source projects have a bar to entry and lots > of potential contributors get left by the wayside. I think Clojure > actually has a pretty good ecosystem around contribution. Could it be > better? Maybe. Could it maintain the high level of quality and still > be more inclusive? Hard to say... > > > * Putting up barriers to entry is \not a good thing. > > I think most people agree with that but the disagreement is on whether > Clojure really is putting up such barriers. I think that's debatable. > > > * Since Lisp is highly extensible, in the long run being > > 'prescriptive' is a losing battle. > > Again, I think this is a debatable point. I don't believe there is any > direct correlation between the prescriptiveness of a language / > framework and its success. > > > * Clojure is already enough of a new way of thinking, and it may be > > simply too much at once for many people. > > Aye, maybe. I don't think there's a gentle path tho'. FP is something > you just have to "grok". In some ways, if you want a middle ground, > there's Scala - a hybrid OO/FP language. I have to be honest and say > that after attending Scala Days recently at Stanford, I was even more > convinced Clojure was the right choice for World Singles (we tried > Scala first). I don't think that FP being "hard" is an issue tho' - OO > was "hard" for people back in the day. > > > * It's meant to be a pragmatic language. > > And I think it succeeds admirably here. We have an incredible > diversity of applications. With Heroku supporting Clojure - and some > great step-by-step tutorials for web applications (with databases) - I > don't think anyone can accuse Clojure of not being pragmatic. I'm > using Clojure for DB work, text file analysis, and all sorts of very > pragmatic, real-world stuff. It's a great language for general purpose > stuff. > > > * The attitude that Smart People are precisely the ones who will want > > to deal with Clojure's existing drawbacks ends up excluding many great > > future Clojurians. > > I flat out don't agree with this on any level. Clojure is pragmatic > and a wide variety of developers from all sorts of backgrounds are > picking it up and building real world stuff with it. It's not just > about "Smart People" nor about any subset of "Smart People". > > > * Final (added) point: while it might have made sense to be > > 'prescriptive' initially in order to establish the identity, core, and > > soul of the language, this has been done sufficiently. Newcomers are > > not going to be confused about what the main points of Clojure are > > now. There is therefore less risk in making it broadly useful to > > different paradigms. > > And I think the new process for moving contrib libraries forward > addresses that. I'm late to the game but folks I trust tell me that > contributing to the new contrib libraries is much easier and much more > open that the process around the old contrib libraries. > clojure.core.unify and clojure.core.logic are testament to serious > innovation within the core Clojure context. > -- > Sean A Corfield -- (904) 302-SEAN > An Architect's View --http://corfield.org/ > World Singles, LLC. --http://worldsingles.com/ > Railo Technologies, Inc. --http://www.getrailo.com/ > > "Perfection is the enemy of the good." > -- Gustave Flaubert, French realist novelist (1821-1880) -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
