"Jamie Brandon" wrote:

The haskell community has a well deserved reputation for being one of the friendliest online communities. Perhaps this would be a good point to figure out what we're doing right? I'm convinced that part of it is that offtopic conversation is encouraged through on haskell-cafe, planet haskell and irc. It makes people seem more human and hence harder to flame.

Well... if you believe what's written in the Cathedral and the Bazaar (which I tend to in this case), which is that the main difference between academic culture and hacker culture lies in the tabooisation/encouragement of personal critique (aka flame, cf. rant), then stuff might be easy to explain. The flame-taboo in hacker culture doesn't actually prevent flames to appear, albeit they are partly hidden and much less prone to escalation than in say alt.politics. The valve tends go get opened only iff you really, really know that you're right and the flamed one doesn't have a chance to fight back with equal force without appearing even more like a fool. Such flames aren't addressed exactly because of the taboo and the respect peers have for the flamer _as a hacker_, not necessarily as a flamer. In academics, OTOH, were asbestos and flame throwers are compulsory, people with highly sophisticated skills in both technical topics and mental warfare are bred. The Haskell community, as I observe it, is a melting pot of both cultures. If now Joe Random Hacker is confronted with a question like "XYZ complains about missing sources, but I have the XYZ package installed", the response won't be "Install a sufficiently leet distro like gentoo or, that failing, install XYZ-dev" (at least without a smiley) because J.R.H. fears to be ripped apart by talk of "co-dependent isomorphic congruency vectors" or something of that ilk... the worst thing being, all that stuff might actually make sense, if you could only understood it. OTOH, academics tend to know that even if you have published n award-winning papers in n+k papers and given n^k talks about them (and have no critics whatsoever, for noone understands what you're talking about), no hacker will even consider to start to respect you if you can't follow up on every one of them with honest-to-model-m, applicable working code. That is, both cultures have their own, diametrically opposite, definitions of bogosity. Nobody wants to appear as bogosity's reincarnation, so everybody shuts up and is nice to each other (except in jest, or, of course, for the hell of it). Does that make sense? /me wonders whether this post is a rant or academics-flame -- (c) this sig last receiving data processing entity. Inspect headers for copyright history. All rights reserved. Copying, hiring, renting, performance and/or quoting of this signature prohibited.

On Mon, 5 Jan 2009 07:27:13 +0100 Achim Schneider wrote:

Achim Schneider wrote:

...

[...]

Hmmm... I seem to have left out the academics definition of bogosity. I bet that others are much more qualified to elaborate on this, but my working assumption has been (and still is) that it does include a wide range of working code, and excludes a lot of abstract nonsense.

Well, as someone who has done research in at least 4 different subfields of computer science, I'd say it depends on which subfield you are in. Academics in Human Computer Interaction might criticise a project for having a badly-designed user interface that is hard to use; academics in formal methods might criticise code for not having a specification anywhere of what it does or how it is supposed to handle erroneous inputs; academics working on aspect-oriented software development might criticise code for "tangling" too many concerns unnecessarily in one function. Whereas, each of these academics might be oblivious (or almost oblivious) to the other two critiques. -- Robin