Computer Progamming Languages

[Tricky Dicky]

Bit of RPG

Wish I'd done some RPG - only for the reason that they're my initials !

 

[Simster]

Anyone that programs in C++ is a professional in my book.

I can understand why you'd say that but have to say that I disagree. I'd say the main advantage of compiling in C++ is the processing speed, and the control you have of memory management which, say, the Java garbage collector otherwise handles for you, sometimes with unpredictable results.

However, I really do think these benefits can actually mask a crap programmer. For example, what good is it saving nanoseconds in processing if you can't write a decent database query (this is often where you want your speed!) or if your app falls over all the time?

 

[Tricky Dicky]

I can understand why you'd say that but have to say that I disagree. I'd say the main advantage of compiling in C++ is the processing speed, and the control you have of memory management which, say, the Java garbage collector otherwise handles for you, sometimes with unpredictable results.

However, I really do think these benefits can actually mask a crap programmer. For example, what good is it saving nanoseconds in processing if you can't write a decent database query (this is often where you want your speed!) or if your app falls over all the time?

If it's low latency you want, the DB should not be on the critical processing path. Certainly I agree you should implement good practices and exception handling though, so your code shouldn't fail, and is handled if it does.

 

[KneeOn]

I started doing Pascal, at Uni this year i've been taught Python and next year i'm doing Web-script programming as one of my optional units.

My discipline in IT though is actually forensics and security - so programming isn't my natural strong point (and i'm frankly shit at it)

 

[Badger]

I did a bit of Amstrad BASIC as a kid but mainly just copying code from a magazine.
The first language I was taught was Visual Basic 6 in college, a bit of a doddle to pick up although it wasn't object orientated (although our teacher liked to pretend it was).
At uni I did bits of Scheme (lecturer seemed to want to teach us something we'd never use), Ada95 (lecturer wrote the language), C and Java.

I now do most of my programming in C# .NET, a very easy language to pick up. I would highly recommend learning some C, it is very powerful and gives you a good grounding for picking up higher level languages (C#, Java etc.) because you get more of an understanding of what is happening lower down. Whatever you do learn I would suggest spending more time learning about fundamental techniques and patterns rather than diving into some code.

 

[father_and_son]

Started with [BASIC, obviously, then] 6502 Machine Code, learnt PASCAL at uni - but remember virtually nothing! -, currently program regularly in VB (all flavours) and SAS and occasionally in Java/Javascript for web based stuff.

I have a passing knowledge of a couple of others but wouldn't profess to being able to code from scratch in them without resorting to nicking code samples from the web!


I find that most of the main languages (All the "C" languages, VB, Java, Perl, Python, etc) these days are so similar that once you have a good knowledge of general programming techniques and a comprehensive grasp of one language, then you can pick up and run with code from any of the others after just a little bit of studying the syntax. Much of the true "art" in programming has been removed since my early days of writing raw machine code back in the late 80s/early 90s.

 

[Tricky Dicky]

I find that most of the main languages (All the "C" languages, VB, Java, Perl, Python, etc) these days are so similar that once you have a good knowledge of general programming techniques and a comprehensive grasp of one language, then you can pick up and run with code from any of the others after just a little bit of studying the syntax. Much of the true "art" in programming has been removed since my early days of writing raw machine code back in the late 80s/early 90s.

There is a huge difference between, say 'C' and C++. OO techniques are a completely different ballgame, though of course you can just write C in a C++ compiler. But I would agree, if you can write C++, then C# or any other OO language should not be a big problem.

 

[alexwright55]

has no one ever used Fortran?

 

[Billy the Fish]

Last point, I detest the number of people who think they can program, either because they can read and maybe alter slightly someone elses code, yet haven't got a clue about code structure, creating re-useable code in library functions, coding standards, etc. Just because someone can copy/paste an example from the web and get something which 'sort of works', that doesn't make them a programmer. There are far too many amateurs posing as 'IT Professionals' IMO.

I'd like to make a point, I detest the number of people who think they can program, either because they can read and maybe alter slightly someone elses code, but they haven't got a clue about code structure, creating re-useable code in library functions, coding standards, etc. Anyone can copy/paste an example from the web and get something which 'kind of works', but that doesn't make them a programmer. I think there are far too many amateurs posing as 'IT Professionals'

Erm, I think that one might have been a bit too subtle

 

[clapham_gull]

Not a traditional language as such, but I'm fascinated with XLST.

Not so much it's ability to transform XML into XML, but's it ability to transform XML into anything.

I also love databases.

So for my masters final project, I decided to transform an XML into SQL.

Why ? Well you can keep the XML quite simple - but transform into complex SQL. Particularly so SQL can incorporate fuzzy logic.

I can read most code (c++, c#, java etc..), but don't develop. Was never really any good at it. Still having to bang away with VBA as much as I hate it.

I do read about languages though. Currently reading a C# book.

 

[thony]

has no one ever used Fortran?

Choice of language should be heavily influenced by the application: for example you wouldn't use COBOL to write games, in the same way as you wouldn't use assembler for MI (at least, not these days).
I started off learning BASIC and Assembler at school, FORTRAN and Pascal at Uni, then trained in COBOL when I started my career. I've then taught myself C, Delphi, and Visual Basic out of pure interest, taken in a few different versions of VB at work, formally trained in Java but didn't end up using it, had a brief go at a couple of Android programming tools, and am now moving onto VB2010 and ASP.NET. My IT career certainly hasn't been standing still!

 

[philsussex]

Not a traditional language as such, but I'm fascinated with XLST.

Me too - XSLT has provided me with hours of fun (how sad!). I had to learn XSLT for work - using it to transform XML to HTML for a on-screen checklists for a Workflow system.

Now with a little easy training, anybody can create new checklists with simple XML which incorporate CSS, regular expressions and Javascript without having to know anything other than a few simple rules.

 

[clapham_gull]

Me too - XSLT has provided me with hours of fun (how sad!). I had to learn XSLT for work - using it to transform XML to HTML for a on-screen checklists for a Workflow system.

Now with a little easy training, anybody can create new checklists with simple XML which incorporate CSS, regular expressions and Javascript without having to know anything other than a few simple rules.

If version 1 supported dates natively, it would be a killer technology !

Once you get REALLY into it, it's incredible what you can do with it. But it's very hard to explain to a traditional developer.

I often use it to create SQL statements, but I often use EXCEL to create SQL statements too.

I come from a different viewpoint I guess. I look straight to the database to learn about an application (if it has one). I'm not that interested in UIs or business logic.

I want to know how the application conceptualises the world around it.

Lost count of the bad databases I'm encountered. I don't want to say that most developers create bad databases, I just wish many would spend more time on it.

BUT..

The other thing I am very interested in indexing unstructured data. I find that fascinating also.

 

[Triggaaar]

Played around with perl a bit, ten years ago, then got sidetracked again

She got about didn't she.

I dipped my toe in to Perl

Cheeky. Did she like it? Did it effect her singing?

 

[beorhthelm]

Purists will argue that the latter two are scripting languages, not programming languages per se...

an odd thing for "purists" (snobs?) to say, as both Perl and Python are full featured programming languages. scripting defines a use case for which these langauges are particularly suited.

 

[clapham_gull]

an odd thing for "purists" (snobs?) to say, as both Perl and Python are full featured programming languages. scripting defines a use case for which these langauges are particularly suited.

There is loads of snobbery around. I heard it in the last few years over an application written in Delphi.

The problem I have most is with "business analysts" who don't understand the development process. They may argue "they don't have to" - but I massively disagree.

I wish that role was elevated and developers were encouraged to move into it, rather than the norm of attractive people people who want to work in IT but never really got the binary.

Of course, not all BAs are like that. I've met some quite brilliant ones. But for every good one, I've met 2 who should really get into something else.

 

[beorhthelm]

The problem I have most is with "business analysts" who don't understand the development process. They may argue "they don't have to" - but I massively disagree.

tell me about it. i was on ISEB course and in a room full of project managers and business analyst and there was only one other technical person. i recall most struggled with the ERDs and one lady didnt have any concept of a database. what i learnt most from the course is that business analysts are taught *not* to think about possible implementation and solutions, just thinking about business requirements and process. i've concluded the solution design is for systems analysts, a distinction i was never able to quiet make before so it was usful. of course the problem is many projects dont get both types of people involved, so either the requirements or the solution gets borked...

 

[clapham_gull]

tell me about it. i was on ISEB course and in a room full of project managers and business analyst and there was only one other technical person. i recall most struggled with the ERDs and one lady didnt have any concept of a database. what i learnt most from the course is that business analysts are taught *not* to think about possible implementation and solutions, just thinking about business requirements and process. i've concluded the solution design is for systems analysts, a distinction i was never able to quiet make before so it was usful. of course the problem is many projects dont get both types of people involved, so either the requirements or the solution gets borked...

Personally, I'd teach all Business Analysts relational theory at least. I'd make it a REQUIREMENT of that ISEB course.

If you can't get your head round an "entity", sorry.. f*** off and do something else.

I really really struggle with a lot of Business Analysts. I've seen loads of fancy "flow charts" which simply aren't appropriate for the technology being employed. The "just make it happen" approach just doesn't work.

In terms of "processes" I've seen it work both ways, badly. Designing a complex process that isn't required because a computer can handle it simply. Or designing a simple process that is computationally very very hard and complex.

Can't get way from it though. A properly old fashioned designed database, or at least a thorough ER process - you are more than half way there.

For the record I am fascinated with dealing with chaos data, documents, the web etc.. But just because the tools are there, doesn't mean to throw away the classic way of doing it. That's isn't clever - that's stupid.

Doesn't need to be complicated - the best solutions are the very very simple ones. Something (again) I struggle to get across to developers.

.. or look if you can buy something FIRST ! .. rather than constantly re-inventing the wheel.

Development SHOULD be the easy bit as long as the analysis is done correctly.

 

[Another Berkshire Seagull]

Did COBOL, PASCAL, BASIC, LISP, Java etc... at university

Been working with C/C++, C#, VB/VB.NET, F#, Javascript in my work over the last 15 years.

Agree with Simster & BTF. IMHO - bad code is in many ways than no code. I work as a consultant visiting many different organisations and shudder at some of the code written that serves as a noose around the businesses neck.

 

[GingerBeerMan]

Thinking of trying to learn C++ as a hobby, but I usually prefer something more web based as the results are usually 'more instant'.

Out of interest what kind of simple programs could I design with C++ (i.e. tictactoe, minesweeper?)