Official Off-Topic Thread

[Gil]

Looking at the examples provided for PHP, Ruby and Python, I find that the PHP example is humanly readable and makes sense. I could read that out loud to a kid and he'd have some sense of what I'm talking about.

Ruby and Python's OO structures make no sense whatsoever. They both have massive advantages over PHP in most areas, but not readability imo. Besides that the synthax of Python always seemed so out of nowhere to me. It's counter intuitive.

I guess it's to each his or her own?

[Ai]

Looking at the examples provided for PHP, Ruby and Python, I find that the PHP example is humanly readable and makes sense. I could read that out loud to a kid and he'd have some sense of what I'm talking about.

I think I should point out -- you don't read out German to an English speaker and judge the sensibility of the Germanic language
based on how much that person understands... that would be measuring something else. IMO a much more meaningful metric is how often the language bites experienced users of it. For example, I find C and C++ bite a lot (C++ more due to added complexity), which seems to match people's common experience of it. (Java is something I haven't tried but seems to bite people at a rate comparable to C++). Python bites very little (meaning I can very often write a routine and it will work right first time) -- this is also confirmed by plenty of other people. I haven't tried Ruby, I would guess it bites a little more than Python (based on the flowcharts I linked). Lua is also pretty good for un-biteyness. Dunno about PHP (it looks like a much simplified Java to me)
Smalltalk seems like it might bite the least of the languages I've heard of, so I hope to try it sometime soon.

Ruby and Python's OO structures make no sense whatsoever. They both have massive advantages over PHP in most areas, but not readability imo. Besides that the synthax of Python always seemed so out of nowhere to me. It's counter intuitive.

I guess it's to each his or her own?

Yeah.. personally, Python/Lua's* OO is the only kind that does make sense to me -- although Smalltalk sounds interesting, I've never actually tried it. C++'s OO is some kind of cthuloid abomination (FYI: I learnt C++ before Python)

I agree that __len__ etc in some senses are not very readable -- otoh __len__ marks it as a 'system' method
(cf.  '_foo' which would be a method internal to the class -- probably only called by other methods to do some low level work)

Intuitiveness is just a measure of how well something fits with your particular habits, according to Peter Sikking (interaction architect working on GIMP UI) and a few other famous usability researchers (eg. jef raskin). So it is indeed personal.

* haven't tried ruby, I'll guess it's probably similar enough.

[Gil]

Personally, I think that what you define as "biting" is just something novice programmers go through, not something that applies to trained professionals who know the difference between memory managed by reference types and dereferenced pointer types (example).

It's all about speed of development really. All languages discussed are mature, potent languages with certain disadvantages and advantages. Simplicity and readability have always been a trophy reserved for C style synthax (C, Java, PHP, C#, etc). Feature set and depth of capabilities traditionally belong to Python style languages (Python, Ruby, others).

The result is that you'll see that the more rigid the project and the more people working on it, the faster companies will go for Java, C++ or C#. Python and Ruby are languages most often used by smaller groups of people looking for a rapid deployment of complex (web) behavior. At least, that's how I feel the business works right now.

If you write a complex codebase in Java, and hand it over to another team, they will easily be able to understand and reuse this codebase (provided you stick to analytic object oriented design, which is a given in Java). If a complex codebase in Python is handed over, things will not always go as smoothly, as the complexity of the language features and synthax tend to obfuscate algorithms, decrease readabilty, etc.

In professional software land, language is not really an issue when talking about one man teams. Choose whatever works best for you. When talking about big teams cooperating (a plethora of programmers working on a codebase over a span of years), you'll see language, platforms, API's becoming very important. The language chosen will depend on the workflow of development of a whole company, which is about complex analys, money, time, personalities, available training to staff, etc.

[blumunkee]

If you write a complex codebase in Java, and hand it over to another team, they will easily be able to understand and reuse this codebase (provided you stick to analytic object oriented design, which is a given in Java). If a complex codebase in Python is handed over, things will not always go as smoothly, as the complexity of the language features and synthax tend to obfuscate algorithms, decrease readabilty, etc.

What?

.....
.....
.....

Wait, what?

[Ai]

Personally, I think that what you define as "biting" is just something novice programmers go through, not something that applies to trained professionals who know the difference between memory managed by reference types and dereferenced pointer types (example).

that terminology was taken from commentary by a professional programmer, who said there were some languages, where you've been using them intensively for years and they *still* give you nasty surprises, and other languages that generally don't. He attributed this to the relative complexity of the offending languages (more precisely, the degree that they fail to be internally consistent and therefore predictable.)

It's all about speed of development really. All languages discussed are mature, potent languages with certain disadvantages and advantages. Simplicity and readability have always been a trophy reserved for C style synthax (C, Java, PHP, C#, etc). Feature set and depth of capabilities traditionally belong to Python style languages (Python, Ruby, others).

C style simple? Perhaps.
Readable? well, it's better than Forth, Perl, assembly, and maybe Lua; I can't really see why you would compare its readability favorably to Python or Ruby.

The result is that you'll see that the more rigid the project and the more people working on it, the faster companies will go for Java, C++ or C#. Python and Ruby are languages most often used by smaller groups of people looking for a rapid deployment of complex (web) behavior. At least, that's how I feel the business works right now.

I can believe that.

If you write a complex codebase in Java, and hand it over to another team, they will easily be able to understand and reuse this codebase (provided you stick to analytic object oriented design, which is a given in Java). If a complex codebase in Python is handed over, things will not always go as smoothly, as the complexity of the language features and synthax tend to obfuscate algorithms, decrease readabilty, etc.

I find it difficult to imagine any knowledgable programmer taking the above claim seriously, although obviously you do.
I think you observed a pattern 'If a complex codebase in Python is handed over, things will not always go as smoothly'
accurately, and then erroneously rationalized it as 'the complexity of the language features and synthax tend to obfuscate algorithms, decrease readabilty, etc.'

In professional software land, language is not really an issue when talking about one man teams. Choose whatever works best for you. When talking about big teams cooperating (a plethora of programmers working on a codebase over a span of years), you'll see language, platforms, API's becoming very important. The language chosen will depend on the workflow of development of a whole company, which is about complex analys, money, time, personalities, available training to staff, etc.

Well, I agree with that, surely.

[Frychiko]

Personally, I think that what you define as "biting" is just something novice programmers go through, not something that applies to trained professionals who know the difference between memory managed by reference types and dereferenced pointer types (example)

Heh.. "professionally trained" programmers? There are such people that get professionally trained to program? Sounds like university graduates who end up not knowing how to program at all.

As for those issues just biting novices.. only a novice would say something like that. Managing memory was a pain in the ass...

[The B.O.B.]

   I'm having a strange problem with Promotion. I wanna' know if someone experienced Promotion users here has ever experienced this; basically, if I try to do a simple 2 frame animation, the 2nd frame ends up being auto cropped/chopped off toward the bottom and right sides AFTER I save it, close it out and reopen it. Doesn't happen if I do 3 frame animations or above. The funny thing is, if I upload that same chopped image, it loads just fine to the web. But if another promotion user saves the file and loads it into his promotion program, than back we go into the chopping issue .
   I've tried re-saving as different file names, saving on to the web, and re-saving it onto my pc, and even reworking it in mspaint, and copying and pasting it...NO difference. I'm totally stumped over this...looked online, and can't find any one who has this problem. I verified my crop settings were normal as well. Any one ever experience this?

[ptoing]

This is very odd. Which version of PM do you use. Also I suggest saving animations and such as .pmp files as that is native to PM and should not yield any problems.

[The B.O.B.]

   I use Promotion V5.1. I was actually able to save it as a ".bmp" file and the 2 frame animation worked just fine. So it must be something with the .gif file type, it doesn't like with 2 frame animations. In either case, I'm going to see if I can email Promotion author, Jan Zimmermann, over the issue. Hopefully he might have a simple solution to this.

Thanks though, Pee-Toe!

[ptoing]

PM6 does this as well. I reckon it's about how gif is stored and how PM is reading it. As in that only changed pixels get stored. It's odd it only happens with 2 framed stuff.