Forget Twitter. COBOL is where it's at.
All the hipsters in Silicon Valley talk about PHP, Twitter, and Web 2.0. But recent surveys show that kids can't be bothered to use Twitter.
Meanwhile, COBOL, one of the industry's oldest programming languages, still "equates to 80 percent of the world's actively used code," according to Stephen Kelley of Micro Focus.
COBOL? Really?
Yes, really. COBOL keeps chugging because, as John Willis suggests, it continues to power the boring (but essential) software like CICS (Customer Information Control System). Not very sexy, but when you think about life for more than a nanosecond, most of what makes life work is the transportation, finance, health care, etc. systems that don't make waves but do make our lives more efficient.
This is why the hot jobs in the cold economy center on "old" programming languages like Java and .Net. They're not cool. They're essential.
I've grown to love Twitter, but I'm not waiting for it to change the world. My demographic (25- to 45-year-olds working in technology) believes it's changing the world, starting with the ushering in of a new age of Iranian democracy. But as Foreign Policy points out, Twitter does as much to help crush dissidents and spread misinformation as it helps to remedy things.
In other words, it's really no different from the old technology, except that it does a better job getting into the news.
In sum, forget the hype. While we think technology runs at breakneck speed, the reality is that technology adoption does not. It's simply impossible, especially for enterprises, to adopt new technology at the pace at which it is developed and released.
This is why, for example, companies pay Red Hat to make Linux move more slowly with an 18-month release schedule. Innovation is great, but many enterprises prefer to get innovation on the drip.
Even when people, usually consumers, do quickly adopt technology, this is often a sign that it will be dropped quickly, too. Consider MySpace. Once on top of the world, it's now hemorrhaging market share to Facebook, which in turn will likely evaporate in the face of The Next Big Thing.
I suspect that the longer the adoption period for technology, the longer it tends to stick around. If I invest a lot of time and money in an IT purchase, I'm less likely to drop it quickly.
I'm not suggesting that Web scripting languages like PHP aren't big, or that Twitter is irrelevant. I'm just saying that the reality of what sells today is often very different from what gets funded today in Silicon Valley.
Linux is now old hat and safe, which is precisely why the value of Linux skills has risen 50 percent in the job market. The same holds true for open source, generally: now that it's old hat, enterprises can't adopt it fast enough.
If you're looking for funding from customers rather than from VCs, you might want to consider that boring, old technology that keeps the lights on but doesn't light up the sky.
Follow me on Twitter @mjasay.
Matt Asay brings a decade of in-the-trenches open-source business and legal experience to The Open Road, with an emphasis on emerging open-source business strategies and opportunities. Matt is vice president of business development at Alfresco, a company that develops open-source software for content management. He is a member of the CNET Blog Network and is not an employee of CNET. Disclosure. You can follow Matt on Twitter @mjasay. 
If your app scales to 100 million users, the alleged differences between "Programming" and "Web Scripting" become much less interesting than the fact that either way you're running highly scalable mission-critical applications.
And I say that as someone who works in large-scale .Net programming, and not as a fan of PHP.
Matt you need a break. you are Hyper confused and even make us confused
Bright and shiny only pays the bills for Start-ups, and VCs. The rest of us have to take orders, build products, manage inventory, and make sure the bills get paid and no amount of chrome makes the engine run better.
I think this article is a much needed self reflection thats desperately needed in Silicon Valley! - which has gotten carried away with its self created sense of the realities and needs of the world!
Who gives a rats behind about Twitter?
If COBOL is running your billing system and your billing system causes you zero issues, you keep it. The problem with COBOL is no one is writing anything *new*, and you can't keep a COBOL maintenance programmer around without work. To a greater extent, the mainframe technology is even better now than Y2K, so I could only imagine what my 15 year old code would be like on modern iron: likely faster, and maybe with some wrappers, even callable from a frontend other than C workstations. It would still process its records faster than anything written in Java or dot-NET, that's a guarantee.
I am surprised no one is taking 2038(I think that is the year) seriously. Lots of embedded devices and legacy software is going to suffer a buffer overflow on UTC since 1970 sometime in that year.
Oh, and it won't be a _buffer_ overflow, that's what happens when an attempt is made to store data ouside the memory range allocated to a buffer. This will be an integer overflow as it will simply be a 1 added to a signed 32Bit integer of 7F FF FF FF (2,147,483,647) resulting in 80 00 00 00 which in a signed integer reads as (-2^31) or -2,147,483,648 which will represent 8:45:52PM, 13th December 1901.
I am aware that the scope of the problem is identical to that of the Y2K bug but that the number and variety of systems affected is much greater due to its affecting current, and not just legacy systems. I'm still not overly concerned. There is still plenty of time to deail with this, it's just a matter of investment of resources.
The reason people want Java and .NET coders over PHP is because they are real languages, applications are complex and often require more than just some server side scripting.
Theory 2: Article was actually written by Sarah Palin.
The article just seemed a little all over the place. Twitter, then Cobol, then old languages like c#, then myspace and facebook... The article is either not focused or I missed the boat. C# is old? Sure like 10 years, what about c++, thats old. or vb6.0 which still has huge support even though MS has killed it. So we migrate onto the latest and greatest to keep up. But like the other guys have said, a php script aint no program man.
Q
http://quintesvanaswegen.com
- by circuit17 August 20, 2009 3:30 AM PDT
- The anti PHP people here clearly don't know much about the language or any software using it beyond basic web pages. They are just regurgitating myths they have read on the internet, or their own prejudices.
- Like this Reply to this comment
-
(30 Comments)I've worked with multiple languages, including Pascal, C, C++, Java, Perl, PHP and a little bit of C#.
Of these my favourites are C++ and C. C++ gets top slot.
PHP is efficient and easy to use, has overall good syntax (being seemingly based on C syntax this is no surprise,) and as of PHP 5 allows proper object oriented programming (PHP 4 offered classes but without private and protected variables, which isn't really the same at all.)
Real programs can be written in Perl, Python, PHP and any other "scripting" language which has a similar level of maturity. The majority of PHP code I write is properly object oriented & modular and generally has little at all to do with output on a web page. I deal with HTML output maybe 5% of the time, if that.
Given a choice between PHP and Java, I choose PHP for ease of use. It's just easier (for me) to write good well structured code using the PHP syntax.
I have seen crappy "script" style code done in Java. I have seen crappy code which doesn't amount to "real" programming in every language I have worked with. Some of the worst code I have ever seen has been in Java and C.
I used to hate Java because 10 years ago it was a lousy and horribly inefficient language (based on comparisons between the same code written in Java and PHP at the time, so it would have fared even worse against a "real" language such as C, C++ or Cobol.) I'm growing up a little now and realising that I don't have enough recent experience with it to be pronouncing judgements.
We all need to learn to equate bad code with bad programmers and not assume a bad programming language.
after all, you can write "web scripts" in C if you want to. :)