CNN is running a story on the
top jobs to have right now. What is interesting is the
responses on slashdot to the fact the .NET is in high demand.
First, there is no argument .NET is the highest demand for developers right now. This is not up for debate, it is fact. You can scratch you head and ask why so many companies are ready to line up behind Microsoft vendor lock in, but it won't change things. Microsoft is masterful at marketing a solution and have created a demand for developers of a very new technology. High demand, low supply, who here remembers their economics 101 class?
Is it the best solution in the market right now? This is an often asked question by developers, and it is irrelevant. Still, .NET 2.0 is a very good platform, so if .NET 1.0/1.1 were lagging 2.0 is making up for it. At the end of the day what a company looks for is the total cost of a platform, or what they perceive that total cost to be. As any developer will tell you, rarely does a company understand all the costs involved.
If a company looked at the true cost of Microsoft, we wouldn't have so many Microsoft shops out there. You need to pay for the server license based on how it will be used. You add in more fees as you add email for your employees. Add in more fees for a database. More fees for developer software. Should you need to add more hardware and servers to handle a growing company, more fees. In reality a Microsoft shop has a great liability over it's head; quick growth could leave the company needing to spend large amounts in licensing and not having access to the capital. Depending on what the company does, the delay in getting funding to grow could kill it.
What about other commercial solutions? IBM, Sun and Oracle all work on the same fees, and their fees are higher than Microsoft's. It is generally accepted that this is a better solution, but the higher costs factor in and cause smaller companies to choose a cheaper Microsoft route. When the small company grows, it becomes hard to leave behind an investment in Microsoft (which includes not only fees, but developed software and IT staff talent, and company data may be locked inside a proprietary format) for another solution.
What about Open Source? First, without a marketing team, most companies may not even understand it enough to know it's a solution. If the company is already a Microsoft shop, then there is still a cost of leaving the current investment. Developers are another problem. You may think it's easy to find open source developers but the truth is open source is flooded with many entry level programmers with little real world experience in anything other than web sites. Languages such as perl and PHP do a poor job at handling large scale and rapid development. Java and python, which do a better job, have lower numbers of programmers; any programmer who can grasp the higher level concepts of these languages and adapt to any language with ease, and so will move to what the market is hiring.
Many perl and PHP developers fail to understand the basic rule of the software developer: learning never stops. There will be new platforms, languages, and technologies so long as there is software. You cannot expect to learn one thing and have it carry you though your career. Not only do you do this to stay employable, you also become a better developer the more methods and approaches you learn. You do this to hone your craft, perfect your art.
If that is something you don't want to do, CNN also says there is high demand for lawyers with a computer science background, so not all is lost.
Posted By Mike On Saturday, February 04, 2006
Filed under developer |
Comments (4)
Daryl
-
Tuesday, February 07, 2006
2:53:00 PM
Ahem, as with java or python, with the right toolkit, PHP's just dandy for rapid development. It scales reasonably well for most things I've ever seen it used in as well, though perhaps for millions of simultaneous users over long times, it wouldn't do so great; even that can be solved with hardware and other software scaling solutions, which are surely necessary to make java and python scalable as well.
Mike
-
Tuesday, February 07, 2006
3:08:00 PM
I knew after I wrote that I would get your comment =p I should have explained what I meant by scale. I don't mean in terms of handling a slashdot or digg - performance wise PHP can hold up, though you better get the Zend engine and not the free version.
Scale in terms of complex design, and multiple developers working on one application. PHP does not encourage a good MVC design - it can be done, but by convention. PHP5 is closer with it's new OOP features, but a far cry from Java and .NET systems where the database is described in an XML file, used by a busniess logic object to model the data that is then used by a user control to interact with it - allowing the page to care nothing of the database and vice versa.
In fairness, python, while further down this road than PHP5, is not at the same point as Java/.NET. Python's Zope is close, but no where near as easy to use.
Daryl
-
Tuesday, February 07, 2006
4:06:00 PM
Why does it matter if MVC is achieved by convention or by built-in structure? I've implemented MVCish applications in PHP even without PHP5. You simply need a generic object that reads table structure to model the data (XML files seem weird for this to me) and you're halfway there.
One of these days, I'll harden and polish up my little MVC thingie (which I playfully call PHP in Pails) and release it to the world so that skeptics like you can see that PHP's not really so crippled a language after all. ;)
Mike
-
Saturday, February 11, 2006
6:39:00 PM
I know I promised you the last word, but I can't resist... PHP in Pails, you have to do it for the name alone!
