I love how the “don’t be a digitial sharecropper” post is followed by the “applications that rely on showing advertising to digital sharecroppers are the future” post.
No, really - google, youtube, myspace, facebook - lovely things, I’m sure, and they don’t get their revenue by making their users pay in cash.
Ther funny thing, of course, is that I don’t buy applications aside from games for my home PC, and I won’t pay for web apps either. But my boss will pay for stuff for work - where “it stops working because the network connection is down or the application vendor retired” are not ideal options and “storing the company’s employee database on a web server” actually makes some people question how thorough an investigation has been made of the vendor’s security qualifications.
The home market has gone for the advertising supported web. They’re not coming back. I suppose they’re the only profitable market, though, so we’ld all better follow.
“WinZip - Web Application Edition” will probably be a market leader.
I do web prog now… but I remember when I was “…not smart enough to do anything else.” except VB… and I put out just as much bad code then as brand-new web progs do now.
Maybe the web programmers in your country don’t use OOP web programming. Maybe only in your country the WP are stupid enough to not draw an UML or a sequential diagram. I’m 99% sure that you have no clue how to build up from ground 0 to the top a big web project. You don’t even know with what problems we confront. I din both desktop applications and web projects. They both have tricky parts, but it’s more easy sometimes in desktop. So read more before opening your wise mouth.
I agree that Web programmers are treated differently to other programmers. I am currently a web programmer/unix programmer. I can code in Php, Perl, Ruby on Rails, Groovy on Grails, use Mysql, configure Apache etc but can I earn the same as C++ programmer. The answer is No!, especially not in the North West of England. But I also know MS C/C++, understand inheritance, virtual functions and have also programmed C code into embedded micro controllers. But still with this experience and more I still cannot earn the same as a C++ programmer. I cant remember the last time I used CSS lol or wrote some HTML but still concidered a web programmer. I think it is a shame that there is such a distinction. If you are a good programmer, have good skills and are keen that should be enough.
Fortunately in the future all apps won’t be web apps. There are apps that can never be web apps. There is and always will be computers that aren’t connected to the internet or any network for that matter. My biggest pet peeve are web apps that aren’t really web apps, they are really desktop apps that run in the browser. Nothing I hate worse then working with a vendor that requires some steep client side run time requirements for their “Web” app. I personally believe that to be a good programmer you need to understand both desktop and web apps and have a strong understanding about the appropriate place to target for a given project. Sadly I don’t see many people with that skill. Too many people thing everything is a web app. My second pet peeve is developers that want to make everything a web service. A lot of times pushing XML around isn’t the best solution, especially when fast response or timely updates are required.
every user-interface will be translated to web for sure. but i would love to see low level applications translated to web apps… it makes no sense to me.
Your article pretty much shows how ignorant you are as a programmer. I write desktop apps and web apps for a living. I can tell you both take the same amount of effort and knowledge to accopmlish.
You are probably just afraid to learn new concepts required for web programming and feel the need to look down and put down those that know how to do it.
Any aspect of programming can be done be morons along with any other proffesion out there.
Think before you speak and right pointless articles.
Michael Braude made some rather foolish comments, but Mr. Atwood follows up with much worse. The general premise is just stupid, and not only that but it is really self centered. I am a web programmer; I can see business apps moving over to the web and think that some sort of web application framework (probably a derivative of the current HTML/Javascript) will probably how the majority of business app UIs operate in the future. So I get how this argument might apply to a small subset of “all programming”. But Mr. Atwood didn’t declare “all UI programming” or “all business app programming.”
What about everything else? All the programmers in the world don’t work on business or social apps! Mr. Atwood, put down your ego for a second and consider the huge code bases I am about to describe:
Operating system: Operating systems for personal computers, for the server infrastructure upon which the web relies, for various network devices such as switches, routers, proxies and firewalls, OS’s for portable devices, for satellites, various other network gear, the DVR, dish network, Tivo, etc, etc, etc. All this code is going to evaporate into Javascript?
Automation systems: All the code that runs factories and other industrial processes are going to move to the web?
Sensitive data systems: All this is going to move to the web? I am talking about the stuff under the covers of banking and other financial systems, national security, local, state and federal law enforcement, etc. I don’t disagree that as encryption and security changes perhaps these systems will become more and more distributed.
The fraking space program.
The world of programming is like an iceberg. Those of us in “application land” are working on that small percentage of the ice above water. There is a giant infrastructure upon which we rest that is under the water. It is my contention that the vast majority of this will never move “to the web”, or at least it will be a very long time until it does.
Its not all about web programming, Mr. Atwood. I really enjoy your blog but sometimes you come up with a blog post that seems truly moronic. I don’t know if you do this simply to spur discussion, but in this case you are completely and totally dead wrong.
Well said Jeff. The web is having this same effect on multiple industries. crowdsourcing is opening the field to people with no formal design training. Blogs, and now book binders, are letting anyone publish their work, and we’ve already seen the music distribution model change dramatically. So why should the developing world be any different.
The thing that people seem to not get is that the more bad writing/design/music/code there is, the more value it gives to the people who are doing these things well. And if you’re doing things well, what really are you worried about?
and as for Michael Braude’s comment “The reason most people want to program for the web is that they’re not smart enough to do anything else”, everyone has got to start somewhere. I’d love to hear what a professional journalist would have to say about his first ever blog post.
Nothing wrong with lowering the barrier to entry so long as you are prepared to have to deal with the poor quality of ‘Product’ that comes as a result.
As you say, the ‘Common People’ couldn’t spell, didn’t know punctuation and had bad penmanship. i.e. the quality of their product wasn’t very good. Fine for their own use but you didn’t see many of them able to compete in the same league as the ‘Pros’.
The web is unreliable by design. An app like word processor must be always ready, so it should stay desktop. So should all ‘professional’ software.
Don’t want your plans to be ruined by failed connection? Stay desktop.
Michael Braude has it backwards. Bad developer produce bad software - they always have and they always will.
The fact the new fad is web apps bears no relation to this underlying truth.
For every high polished piece of software there are twenty truly awful piece of software. It doesn’t matter if that’s a web applicatio or a desktop application.
Has Michael really never used a crappy, half-baked, desktop application? Why would he think the web is any different?
I won’t say that your post is wrong, but neither will I say Michael’s is. I specifically agree that if there was ‘only’ web programming left, then I would probably leave the profession.
But of course, you’ve forgotten these ‘niche’s’!
Web servers
DNS servers
Database servers
Mail servers
Operating systems
Web browsers!
the list goes on…
I for one am happier working on these systems than doing the same thing over and over in asp.net. Sure… you can do something new and original (like google maps) with the web, but how likely is it that the company you are working for is creating something that cool? More likely they are creating some shoddy website or b2b software asp.net front end, sql backend, needs to be done by september. No thanks!
And there are plenty of “bad engineers” in desktop development too. So the barriers of web development are lower, and consequently a larger volume of developers follow. This attracts a high volume of sub-par engineers, but what makes him assume that it’s proportionally higher? He fails to consider that with a crowd of sub-par engineers also comes a higher volume of top-notch engineers.
To say that the “Internet is responsible for a collective dumbing down of our intelligence” only really indicates his arrogance in how exclusively important he thinks one type (and his type) of development is.
"As I see it, there are two major things new developers need help understanding:
How to write good, clean code.
Why it matters.
(Their managers also need help understanding #2.)"
No. Managers need to stop hiring inexperienced self taught developers. My shop is not your school. My code base is not your final exam. How many basic concepts might you still not be aware of? I will help these inexperienced developers with understanding the things you mention by directing them to the nearest educational facility.
Glenn, actually even if you write a spanking 3D-rendered desktop program for highly responsive and interactive experience, any action that retrieves data or requests an action from some server out in the Internet is still likely to generate some lag.
This is still the sad nature of network computing on the Internet scale today - too many routers/hops taking too long to pass your data packets back and forth.
