Programming languages trends (TIOBE)
Locked
Programming languages trends (TIOBE)
 
|
http://www.tiobe.com/index.htm?tiobe_index
Smalltalk is on 33 position with 0.133%. Who cares? How to promote? |
Re: Programming languages trends (TIOBE)
|
|
(long email)
I will pick up on this thread. I care, because I believe in that Smalltalk makes me more productive. :). While at first sight we can see the limited spread of Smalltalk as nothing else but a huge disadvantage, I believe there are some point of views that can turn it into an opportunity. 1. In a competitive market, everyone is looking for the competitive edge, so if we believe that Smalltalk gives us that edge we should be happy to be part of the privileged ones that can exploit this power :). 2. Still, a larger community offers a better safety for the effort investment. At the moment, we are among the 0.133% of programmers. Very low, but we get by. Now, if 0.133% is not much then it also should not be hard to double it. And if we double it, then it should get twice as good as it is now. And as I said, right now we get by Ok :). Of course, I am exaggerating a tiny bit. If we want to enlarge the Smalltalk community, we need some marketing efforts. From this perspective I love what Smalltalk bloggers are doing, and in particular James. But the fight is hard mostly because the greatest enemy of Smalltalk is its own name. This name carries an old story of how this name used to be something and how it is not anymore. It does not matter how we got here, but the fight for repositioning a "dead" or "old fashioned" product is very hard. So, I would say that the first message we should send is that Smalltalk is alive and kicking. And the best way to do that is to offer tools/gadgets that can appeal to non-smalltalk developers, so that they have an incentive to download them, and like that to see that Smalltalk is not dead. These tools/gadgets must be free and readily usable. Seaside is probably the flagship here and I love it that every Smalltalk platform embraced it. But, Seaside still requires one to actually use Smalltalk. Better are products built on top of it that do not require Smalltalk upfront. For example, Pier would be such kind of a product on top of Seaside. BottomFeeder is also an example. Moose is another example. The nice thing about Moose is that it appeals to people that program in other programming languages and they tend to want to use it, and like this they learn that Smalltalk is not dead. Cheers, Doru On Feb 2, 2008, at 10:48 PM, Vladimir Pogorelenko wrote: > http://www.tiobe.com/index.htm?tiobe_index > > Smalltalk is on 33 position with 0.133%. > > Who cares? > > How to promote? > -- www.tudorgirba.com www.tudorgirba.com/blog "Every thing should have the right to be different." |
Re: Programming languages trends (TIOBE)
|
|
Tudor Girba wrote:
> > But the fight is hard mostly because the greatest enemy of Smalltalk is > its own name. This name carries an old story of how this name used to be > something and how it is not anymore. It does not matter how we got here, > but the fight for repositioning a "dead" or "old fashioned" product is > very hard. I quite strongly disagree with this analysis, the main problem with Smalltalk is that its development paradigm is so different from mainstream dev processes, not its age. In the commercial arena it is viewed as a huge plus that there are Smalltalk deployments that have been upgraded over decades and that Smalltalk itself has been refined over decades instead of being some newfangled big bang development. I think we should cultivate the view that Smalltalk is a venerable language. It is field-proven. I can well understand that the geek culture/amateur users regard Smalltalk as 'dead' but I don't think that is induced by Smalltalk itself. Rather all the lovely stuff that happened to computing before the 90's (theory formation, new paradigm creation, whatever) mostly landed in a huge 'black hole' in the collective consciousness. Since this is a problem that is separate from Smalltalk I don't think it can be mitigated by rebranding Smalltalk itself. This point was driven home in a rather unpleasant way when I read the recently published book 'Beautiful code', a collection of essays by different authors each pointing to the beauty they perceived in some code or system or design. Frankly most of them were full of sh*t, I couldn't reconcile their story with my sense of beauty, most of them sounded like silly young kids advertising their latest hobby project. Indeed, the only essay that I actually liked in that book was really about 'ugly code', it described the development of a correct(!) priority inversion management algorithm for some OS, that essay alone made the book worth its money. Another illustration is the TeaTime protocol used in Croquet, it was published several decades ago but never got implemented until the Croquet project came along. The software development community as a whole is very poor at reusing existing knowledge. > http://www.tiobe.com/index.htm?tiobe_index > > Smalltalk is on 33 position with 0.133%. > > Who cares? > > How to promote? > This index is based on web searches so there are two obvious tracks to persue: 1) Determine how compatible Smalltalk vendor's websites are with their ranking algorithm. It might well be that we don't care enough about web presence since we get most of our jobs through other channels (word of mouth etc) - the internet might not be as important for us as it is for newer languages that are centered around a geek culture. My assumption is that geek culture networks through the internet while the commercial culture networks through more personal interactions - so the algorith is skewed against us. 2) game the system :-) R - |
RE: Programming languages trends (TIOBE)
|
|
Hi Reinout,
You are generally right, as always. But I don't think that Tudor was saying that Smalltalk's age is the problem. He said that the problem (in his opinion) was ' an old story of how this name used to be something and how it is not anymore'. Whether that's correct or not is another story. Ivan > -----Original Message----- > From: Reinout Heeck [mailto:[hidden email]] > Sent: Thursday, February 21, 2008 1:11 PM > To: vwnc > Subject: Re: Programming languages trends (TIOBE) > > Tudor Girba wrote: > > > > > But the fight is hard mostly because the greatest enemy of Smalltalk > > its own name. This name carries an old story of how this name used to be > > something and how it is not anymore. It does not matter how we got here, > > but the fight for repositioning a "dead" or "old fashioned" product is > > very hard. > > I quite strongly disagree with this analysis, the main problem with > Smalltalk is that its development paradigm is so different from > mainstream dev processes, not its age. > > In the commercial arena it is viewed as a huge plus that there are > Smalltalk deployments that have been upgraded over decades and that > Smalltalk itself has been refined over decades instead of being some > newfangled big bang development. I think we should cultivate the view > that Smalltalk is a venerable language. It is field-proven. > > I can well understand that the geek culture/amateur users regard > Smalltalk as 'dead' but I don't think that is induced by Smalltalk > itself. Rather all the lovely stuff that happened to computing before > the 90's (theory formation, new paradigm creation, whatever) mostly > landed in a huge 'black hole' in the collective consciousness. Since > this is a problem that is separate from Smalltalk I don't think it can > be mitigated by rebranding Smalltalk itself. > > This point was driven home in a rather unpleasant way when I read the > recently published book 'Beautiful code', a collection of essays by > different authors each pointing to the beauty they perceived in some > code or system or design. Frankly most of them were full of sh*t, I > couldn't reconcile their story with my sense of beauty, most of them > sounded like silly young kids advertising their latest hobby project. > > Indeed, the only essay that I actually liked in that book was really > about 'ugly code', it described the development of a correct(!) > inversion management algorithm for some OS, that essay alone made the > book worth its money. > > Another illustration is the TeaTime protocol used in Croquet, it was > published several decades ago but never got implemented until the > Croquet project came along. The software development community as a > whole is very poor at reusing existing knowledge. > > > > > http://www.tiobe.com/index.htm?tiobe_index > > > > Smalltalk is on 33 position with 0.133%. > > > > Who cares? > > > > How to promote? > > > > This index is based on web searches so there are two obvious tracks to > persue: > > 1) Determine how compatible Smalltalk vendor's websites are with their > ranking algorithm. It might well be that we don't care enough about > presence since we get most of our jobs through other channels (word of > mouth etc) - the internet might not be as important for us as it is for > newer languages that are centered around a geek culture. My assumption > is that geek culture networks through the internet while the commercial > culture networks through more personal interactions - so the algorith is > skewed against us. > > 2) game the system :-) > > > > > R > - |
Re: [vwnc] Programming languages trends (TIOBE)
|
|
Indeed, I am saying that the enemy is how Smalltalk is perceived at
the moment. And to my mind it is perceived as something that used to be, and that is not "modern" anymore. I also dislike the idea that only "new and modern" things are good, but this is how the world (at least the one I go around into) tends to think these days. Of course, we can choose to say that the rest of the world is narrow minded, and that they reinvent the wheel all the time. But the thing is that even though we would be right, we would still not make them look at Smalltalk like that. Reinout says that in the commercial arena Smalltalk is regarded as a huge plus. Maybe I was just unlucky, but in the companies I visited Smalltalk does not exist. Or if they know of it, they know of that technology the previous team migrated to Java some years ago. As I said, I strongly believe that the Smalltalk is many respects better than other languages I know. But this is not a battle of inner quality, it is a battle of image. When it comes to convincing someone to choose Smalltalk, it's not about what we know about Smalltalk, it's about what that someone knows about Smalltalk. And if we want to be effective at convincing, then we have to speak their language. If I am right, and people do think that Smalltalk is dead (if they know of it at all), then this is the very first thing we need to fight against and bring the message that Smalltalk is alive and kicking. Of course we can tell them that Smalltalk is alive, but I think a much better strategy is to actually give them something kicking that they want, and then tell them that it is done in Smalltalk. Cheers, Doru On Feb 21, 2008, at 6:39 PM, Ivan Tomek wrote: > Hi Reinout, > > You are generally right, as always. But I don't think that Tudor was > saying that Smalltalk's age is the problem. He said that the problem > (in > his opinion) was ' an old story of how this name used to be something > and how it is not anymore'. Whether that's correct or not is another > story. > > Ivan > >> -----Original Message----- >> From: Reinout Heeck [mailto:[hidden email]] >> Sent: Thursday, February 21, 2008 1:11 PM >> To: vwnc >> Subject: Re: Programming languages trends (TIOBE) >> >> Tudor Girba wrote: >> >>> >>> But the fight is hard mostly because the greatest enemy of Smalltalk > is >>> its own name. This name carries an old story of how this name used > to be >>> something and how it is not anymore. It does not matter how we got > here, >>> but the fight for repositioning a "dead" or "old fashioned" product > is >>> very hard. >> >> I quite strongly disagree with this analysis, the main problem with >> Smalltalk is that its development paradigm is so different from >> mainstream dev processes, not its age. >> >> In the commercial arena it is viewed as a huge plus that there are >> Smalltalk deployments that have been upgraded over decades and that >> Smalltalk itself has been refined over decades instead of being some >> newfangled big bang development. I think we should cultivate the view >> that Smalltalk is a venerable language. It is field-proven. >> >> I can well understand that the geek culture/amateur users regard >> Smalltalk as 'dead' but I don't think that is induced by Smalltalk >> itself. Rather all the lovely stuff that happened to computing before >> the 90's (theory formation, new paradigm creation, whatever) mostly >> landed in a huge 'black hole' in the collective consciousness. Since >> this is a problem that is separate from Smalltalk I don't think it >> can >> be mitigated by rebranding Smalltalk itself. >> >> This point was driven home in a rather unpleasant way when I read the >> recently published book 'Beautiful code', a collection of essays by >> different authors each pointing to the beauty they perceived in some >> code or system or design. Frankly most of them were full of sh*t, I >> couldn't reconcile their story with my sense of beauty, most of them >> sounded like silly young kids advertising their latest hobby project. >> >> Indeed, the only essay that I actually liked in that book was really >> about 'ugly code', it described the development of a correct(!) > priority >> inversion management algorithm for some OS, that essay alone made the >> book worth its money. >> >> Another illustration is the TeaTime protocol used in Croquet, it was >> published several decades ago but never got implemented until the >> Croquet project came along. The software development community as a >> whole is very poor at reusing existing knowledge. >> >> >> >>> http://www.tiobe.com/index.htm?tiobe_index >>> >>> Smalltalk is on 33 position with 0.133%. >>> >>> Who cares? >>> >>> How to promote? >>> >> >> This index is based on web searches so there are two obvious tracks >> to >> persue: >> >> 1) Determine how compatible Smalltalk vendor's websites are with >> their >> ranking algorithm. It might well be that we don't care enough about > web >> presence since we get most of our jobs through other channels (word >> of >> mouth etc) - the internet might not be as important for us as it is > for >> newer languages that are centered around a geek culture. My >> assumption >> is that geek culture networks through the internet while the > commercial >> culture networks through more personal interactions - so the algorith > is >> skewed against us. >> >> 2) game the system :-) >> >> >> >> >> R >> - > > -- www.tudorgirba.com www.tudorgirba.com/blog "One cannot do more than one can do." _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
Re: [vwnc] Programming languages trends (TIOBE)
|
|
Tudor Girba wrote:
> > If I am right, and people do think that Smalltalk is dead (if they > know of it at all), then this is the very first thing we need to fight > against and bring the message that Smalltalk is alive and kicking. Of > course we can tell them that Smalltalk is alive, but I think a much > better strategy is to actually give them something kicking that they > want, and then tell them that it is done in Smalltalk. > In geek society Smalltalk doesn't seem to exist. Current memes that I've seen floating around are that JUnit is the first unit testing framework, refactoring was invented by Microsoft etc. So in that respect we have a lot of our territory to steal back, before we can advance. I'm guilty of that too, I keep yapping in our inner circle but don't blog, don't write articles etc. I've seen similar tendencies in the Forth community. Maybe holds for all programming languages that are 'very good'? If we are so bad at evangelizing Smalltalk publicly, maybe we can choose our targets more strategically: don't just evangelize interested geeks but have a look at our personal networks and specifically target authors/publishers? Let /them/ do the writing if we are so bad/lazy at it... R - _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
Re: [vwnc] Programming languages trends (TIOBE)
|
|
In reply to this post by Tudor Girba-3
I won't give a dime for those "language shootouts" and hit lists.
Objective-C is at a mere 0.0034% (? site is currently down). Nevertheless, the Cocoa lists at Apple are extremely active and vibrant with a couple hundred posts daily. Although the Cocoa framework is being modernized and extended on a continous basis, the OpenStep, NeXTStep and whatnot behind it is really old. Tudor Girba wrote: > Reinout says that in the commercial arena Smalltalk is regarded as a > huge plus. Maybe I was just unlucky, but in the companies I visited > Smalltalk does not exist. Or if they know of it, they know of that > technology the previous team migrated to Java some years ago. > I would guess these ports are more likely to happen to web and intranet projects where Java is considered the politically correct choice. There are other projects where Java is /not/ an option and where Smalltalk (especially VW) greatly outperforms other platforms. > ... I think a much > better strategy is to actually give them something kicking that they > want, and then tell them that it is done in Smalltalk. > That's what Seaside can do. Seaside rocks. Andre _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
«
Return to VisualWorks
|
1 view|%1 views
| Free forum by Nabble | Edit this page |