On Fri, Jan 27, 2012 at 1:53 PM, Guido Stepken <gstep...@googlemail.com>wrote:
> I see quite a difference between "doing things right" and "doing the right > things" ! :-) > Agreed. Don't let the perfect be the enemy of the good and all that. But we're talking at different levels here. I want to hear what Marcus thinks to my reply to his post. That's where this thread comes from. > Am 27.01.2012 22:50 schrieb "Eliot Miranda" <eliot.mira...@gmail.com>: > > >> >> On Fri, Jan 27, 2012 at 11:51 AM, Guido Stepken >> <gstep...@googlemail.com>wrote: >> >>> Hi Elliot! >>> >>> When I rethink, why new programming languages came up from zero to a >>> significant market share, like PERL, PHP, Python, Ruby, JAVA, C# (.net) >>> Visual Basic, Visual C++ and others died out, like Delphi, >>> TurboBasic/Pascal/C I could name different reasons: >>> >>> - Free license vs. expensive >>> - Wrong payment model (per developer, per runtime, both) >>> - Good, free support on websites vs. "Bronze/silver/gold" >>> paystupid-support >>> - Attractiveness of one "killer app" that made programmers change to >>> another language >>> - Portability of code onto other platforms >>> - Mightyness of libraries >>> - Missing standards, protocols, support of hardware >>> - Good vs. bad marketing, deciders not convinced that product will >>> survive/missing timeline, visions, lack of money in background >>> - Subcritical mass of programmers using product, lack of professionals >>> >>> That was in former times. >>> >>> Today, new criterias play a far more relevant role, hat haven't really >>> existed just 3 years ago: >>> >>> - Has it (the OS,the programming language and GUI framework) an >>> appstore/plugin concept to let free, creative brains being able to >>> participate, earn money with? >>> - Barrier - free payment model included (mobile payment, card, bank >>> account)? >>> - Free use with sponsoring by ads possible (programmers payed from >>> multiple resources, not user alone) >>> - Cryptographic prevention of missuse included? >>> - Free and matured SDK available? >>> - Connections to social software like facebook/twitter/Google+/Groupon >>> included (API access, programming language and all protocols supported) >>> - GUI designed for desktop as well usable for touch and self adapting to >>> different screen/touch sizes? >>> - Touch gestures possible and lib avail? >>> - Microsofts Kinect hardware/video recognition of faces, hand/face mimic >>> gestures possible and supported in libs? >>> - Voice recognition supported? >>> - Mobile ready? (touch, GPS, compass, barometer, gyro, hardware OpenGL) >>> - Rockstable? >>> - Fast, running in low power devices? Joule per clock cycle ratio??? >>> - Critical mass of users already reached, increasing? >>> - Critical number of apps there to raise interest? >>> ... >>> >>> So, the Pharo developers might now decide, what to invest their >>> brainpower into! :-) >>> >>> Just my 2ct. >>> >> >> OK, that looks like a great list. But don't you agree that criticism (in >> the sense of something that leads to quality software engineering) >> underlies several of these, such as Rockstable, Fast, running on low-power >> devices, etc? To me, being critical doesn't mean being uncreative or >> conservative; it means thinking about what you're doing, and doing a good >> job. >> >>> Guido Stepken >>> Am 27.01.2012 19:46 schrieb "Eliot Miranda" <eliot.mira...@gmail.com>: >>> >>> >>>> >>>> On Fri, Jan 27, 2012 at 5:33 AM, Marcus Denker >>>> <marcus.den...@inria.fr>wrote: >>>> >>>>> >>>>> On Jan 27, 2012, at 6:13 AM, dimitris chloupis wrote: >>>>> >>>>> > This article is really encapsulates the attitude and what is wrong >>>>> with programming in general. The attitude of superiority and intelligence >>>>> that seems to plague coders and being the biggest obstacle to progress. >>>>> >>>>> Yes! The "Everyone is dumb but me" phenomenon... >>>>> >>>>> What those "intelligent" people don't get is that complexity is >>>>> inherently exponential. So even if you are >>>>> 10 times more intelligent than me (very well possible), it is >>>>> *completely* irrelevant considering that complexity >>>>> grows non-linearly. >>>>> >>>>> If you combine this with the notion of Evolution: that it is >>>>> impossible to creat "the perfect" out of nothing, yet >>>>> entropy grows when you incrementally improve things... than this has >>>>> some very serious consequences. >>>>> >>>>> > For me the main problem with is the whole aura of "elitism" , what >>>>> better example than Lisp, where beginners are attacked and be excluded. >>>>> >>>>> We had the same effect in Squeak at the end. No progress, every >>>>> improvement was actively fighted against, if needed with the nice argument >>>>> that >>>>> one can do it even better, and only "the best" is worth for Squeak. >>>>> >>>>> Another thing that "intelligent" people don't get is that critizising >>>>> is trivial: You can *always* do better, there is no perfection. It's an >>>>> endless process. >>>>> This implies that one has to accept and embrace imperfection if one >>>>> wants to have a future. Else you end up never finishing anything, the >>>>> death >>>>> of any >>>>> incremental progress. >>>>> >>>> >>>> But criticism is essential. How does one identify a mistake if not by >>>> criticising? There's a huge difference between constructive criticism >>>> (analysis, testing, comparison, evaluation, measurement) and negativity >>>> (denial, fear, slander). How can one engineer without measurement, without >>>> thought? Being agile doesn't imply being random. Evolution measures, and >>>> most harshly; the weaker don't survive. >>>> >>>> >>>>> Pharo was started with the explicit goal to do as many mistakes as >>>>> possible, as fast as possible. >>>>> >>>>> Marcus >>>>> >>>>> -- >>>>> Marcus Denker -- http://marcusdenker.de >>>>> >>>>> >>>>> >>>> >>>> >>>> -- >>>> best, >>>> Eliot >>>> >>>> >> >> >> -- >> best, >> Eliot >> >> -- best, Eliot
