.Net lies

From:
http://www.webmasterbase.com/article/1005/

And there was the article reading .NET speed over Java. Further M$ will bring out new versions of office with propertiery XML formats that .NET *will* integrate with.

I'm willing to bet that freebsd (or linux)/postgres/php can better .net in a number of ways, the most important being they are open standards, have better integration of products (templates, xml support, apache integration, srm, etc...), better support, easier to learn and more efficient.

What are peoples thoughts on the .net marketing promises?

Z.

PS> That should be .Net Lies "?" ie do you think they are telling the truth or not ???

PHP is a programming language with a heavy emphasis on rapid and efficient web development. .NOT err.. .NET is a programming framework with a focus on making Windoze programming suck a little less by bundling a bunch of Unix philosophy into some cute buzzwords and selling that bundle at exhorbitant rates. You can't really compare the two, so I don't know where Bozo McGee in that article came up with that idea (Ok... so maybe I should RTFA... but I don't feel like it ).

However! You CAN compare PHP to VB or VC or J# or C# or F flat maybe.... and it whoops *** there when you compare it on the merits of web development only (try comparing it on the merits of Windoze GUI development to see the results flip-flop... but hell.. who in their right mind develops on Windoze anyway? ):

1. PHP is a fairly portable language. *.NET languages are not at all portable, even between different Windoze versions in many cases.

2. PHP and MySql will absolutely whip anything Microsoft / ODBC driven - including Sql Server. At least from what I've seen in the way of benchmarks (of course, I don't link to benchmarks because they are largely, IMHO, bull****).

3. PHP / Apache memory management is far superior to most ASP/IIS mm cases (especially IIS4 and under). Disclaimer: I never followed up on this.. the latest versions of IIS may have fixed a lot of the memory mgmt problems / leaks / bugs... but my money's on not.

4. PHP is free. Now, granted, you can download and install the .NET framework for free.... if you have Windoze Xp or 2000 (ka-ching - ante up at least $350 for that free framework). However, to get Visual Studio.NET, you'll add at least another $999. From there, you can peicemeal a decent development system together by purchasing further ASP extensions. Ouch.

5. PHP has a helpful, open community of developers to fix bugs and provide discussion and support. I don't know if you've ever tried to get "support" from M$ for anything... but you'd be better off buying a box of fortune cookies than wasting your time with it.... oh.. of course.. you can buy the "Wide World of Incompetence" CD that catalogs all the bugs, security holes, problems, and general headaches of the Windoze world - for a hefty price.

Interestingly.. I believe that Perl and Python (probably through ActiveState) both have plans, or already do, support the .NET framework.... can anyone confirm?

Sorry.. this post was not well-thought out and is largely a matter of personal opinion with a smattering of useless factoids... but still.. I'm not lying about anything there (or, at least not trying), and I think just from those items I mentioned above, PHP is obviously superior.

... a perl programmer defending PHP from an article he didn't read... HERESY!!

Hey Z,

I got a good laugh over that "superb class library" example in the SitePoint link you posted. Gee, I dunno... I must be missing something, because I have been using class libraries in PHP to manipulate HTML for years... And, my classes usually require a lot less fuss than that example.

Here's the thing about so many of these discussions that compare one programming environment to another: the people making the comparison always compare the apples from one system to the oranges from another.

Yes, PHP allows you to use simple stupid procedural PHP with inline HTML. Any PHP developers who stay with that approach for more than a year should loose their geek license. There are great abstraction toolkits for HTML. In fact quite a few good ones to choose from.

Reminds me of another discussion with someone who wanted to show how great it was that you could do ASP.NET stuff without mixing HTML and ASP. Wow, wonderful. Again, something I have been doing with PHP templates for years. The syntax is slightly different but the result is the same.

And PHP even has an "official" HTML class, in Pear, which can probably accomplish everything ASP's HTML class can, with considerably less fuss, and (I'm sure) much better cross-browser compatibility.

.NET lies? I don't know. I really don't care. I just don't like the idea behind .NET that much. To me it seems like a specious argument. Looks good on the surface, with ideas of a Common Language Runtime, etc... but it sounds like a big steaming mess is about to hit the fan.

C#, on the other hand, taken by itself, seems interesting. If we get a good implementation of C# for Linux and FreeBSD, I wouldn't mind checking it out. From everything I hear, the C# language is a nice, sensible, well-designed language.

But my one experience with .NET so far ended in utter failure. I was simply attempting to install the trial version of Dataphor on my Windows 2000 Pro system. The guys at Alphora invented a great-looking relational DBMS and toolkit, but for some reason decided it had to be done in .NET. So, wanting to check it out, mainly for curiousity, I set out one day to get the .NET runtime on my system and then to install Dataphor. After literally a whole day of downloading service packs, rebooting, following incomprehensible instructions (do things in the wrong order and you are toast), I still could not get even a simple ASP.NET page to run. After another half day of tweaking and reading, I finally decided this was not worth my time. And they say open source stuff is difficult? In less than half a day I can have a full installation of FreeBSD, with a custom kernel, Apache, PostgreSQL and PHP all compiled from source with my favorite options, and SSL/SSH for security. And this is if I install FreeBSD remotely, over FTP!!

So, if this is how hard it is to get it to run properly on Windows 2000, I think we are in for a long hard road to turn it into something even remotely resembling write-once, run everywhere.

Personally, I am hoping XUL will obviate much of the perceived need for .NET. But that might just be one solitary lone geek's dream... .

Thanks guys, and rofl R You'all make me feel better.

I see .NET is M$ strategy to attempt to catch up on the lost ground it percieves for the web service movement. I heard C# was a copy of java, and I know at least here, Java is seen as M$ biggest competitor. Along with .NET came a number of damining comparisons like the famous pet shop comparison. However like you point out R, .NET sounds good but in reality it is trying to be all things to all people.

I see M$ releasing new version of Office with a proprietary XML based format. However .NET developers would be able to access this format easily and integrate "document management" principles with "content management" or more simply, Intranet and Internet. I don't believe that this will work that well, because they are taking a product that isn't able to integrate with itself and creating something like Open Office which uses an Open Standard XML/MathML for storing writer (word), calc (excel) and impress (powerpoint). [No smeg, rename a OOo file as .zip and unzip it, check out content.xml - how is that?]

I think that when you build systems/products/services around standards you find that you can easily use them together. For example how well do PHP and Apache work together? Add in FreeBSD (or Linux: Debian/Redhat/etc for that matter) and Postgres and you have a very powerful combination. Now imagine if apache worked on a php copy... and a bsd copy ... and a database, how much time would they have to ensure that apache was the best webserver in the land? Ever wonder why IIS is so far behind Apache?

The biggest hurdle that I see for .Net and M$ is security. They products weren't built to be secure. And hear the recent MS SQL worm (sapphire) that exploited a known bug and wiped out databases across the land? Or worms spread through express/outlook or windows or IIS? And the amount of service patches required? No way buddy, I don't have time for that, especially when I could run bsd/apache/postges have all the security I could ever want and never have to touch the beast with service packs etc.. etc....

Now Java is a cool alternative. But it isn't for me. It simply doesn't have as cool and supportive community. I went and checked out that class R and the rest of pear at http://pear.php.net/packages.php. How awesome is PEAR, and whole library of classes that integrate and too so much from XML, SOAP, HTML, HTTP ... If you've ever thought you needed something from PHP you can be sure that somebody is working on it somewhere and will not only release it free, but support it properly as well. This is more that we can say for M$ their support is uniform: "just upgrade" and if that doesn't work "you're on your own." THis was my experience and I'm sure they are not always that bad, but never will they be able to compete with open source. Why? Because Open Source is a community where we all help each other. There are people making money like Zend and Suse, and their are products supported by Sun (postgres and OOo) but it is the people developing the systems and us using them that make this difference.

Cool so now how to make all these businesses in New Zealand understand that we use Open Source and it gives us an advantage over .NET? This is what I need to do, because everybody expects .NET. Here's what I have:

Support
*Open Source provides great real support for its products
Quality and Standards
*Open Source aims for clear standards to make life easy for developers (consider XUL ) as well as quality to ensure bugs are tracked and fixed quickly
Community
*Open Source provides large code llibraries for developers to use to save time
Cost
*Open Source is free, although if corporates want to pay for commercial support I'm sure that Zend, Postgres evern VL-SRM would have no problems

Please add/comment because it is difficult to get work down here right now, and I need to get this Open Source thing pumped!

<Note that not all OS projects fit into these categories, obviously, but I'm thinking of R's nirvana: BSD/PHP/Postgres etc>

Thanks again all (yeah just everybody who hangs at devshed - you rock), and especially Ctb and Rycamor for your help

Z.

Just a quick note, Z:

Remember also that if you combine FreeBSD/Apache/PHP/PostgreSQL, each of those has a BSD-style license. Ergo.... there is nothing to stop a commercial company from rolling them all up into a nice package and selling it, support and all.

That's what I think is missing in this picture. You see, the big businesses interested in open source have almost all focused on Linux, which is GPL, thus not allowing full commercial integration. I was just reading an interview with Sun's Scott McNealy, which focused on Sun's "Linux strategy". (All the big companies have a "Linux strategy" now. It's like the hot new thing, and if you don't have one, you are so out of it)

But anyway, the interesting part came when McNealy mentioned what he perceived as Sun's advantage in this market, which was the fact that they make their money in hardware. He said something like "Unlike the other companies, we can give away Apache, we can give away Postgres, and we can still make money".

So, while discussing Linux (GPL) strategy, he mentions two BSD-licensed apps--including Postgres, of all things.

Start connecting the dots, boys and girls: the only other piece of the puzzle Sun needs to get straight is FreeBSD, not Linux, and they have a software package for their machines. I wonder when it will hit them. Or if ?

But anyway, all it would take is for the right company to get this idea, and they could take this anywhere. Not a super huge slow-moving company, but a nice medium-sized tech company, with some respectability behind them. Get them to throw their weight behind this sort of concept, and that might make a big splash.

And don't forget that Apple is delving heavily into BSD-land itself. They just might be the right company to take this on. Check this out: my rich uncle just bought a new top-of-the-line G4, with the cinema display. I sat down, having never really played with OS X before, and thought "Hmmm... just how on-board is Apple with all this". I found the command shell, started checking a few things out, and within literally 5 minutes, I had a full-blown Apache/PHP system running. It was already installed, just waiting for my techie curiosity to configure it and start it up. And don't forget, Apple is trying to get into the server market, too.

Hmm... again, start connecting the dots, kids. It's happening sooner or later anyway. That's the good news . Be ready.

Yo, Z!! Check out this Slashdot story. Sounds like you need to join up with the New Zealand OpenSource Society .

Thanks R, in my own backyward too how rlof. I've joined up and will let you know if anything cool comes out of it.....

Awesome,
Z.

Speaking of .NET lies, it's coming true sooner than I thought. Now Microsoft is patenting all the supposedly "open" interfaces to .NET. So much for Mono, .GNU, etc...

I just love that title MS patents .Everything

Thanks for sharing and I am not suprised, just wait for the closed XML move

Z.

Hey R,

This article supports "kind-of" what you're saying above:

http://news.com.com/2010-1071-984072.html?tag=techdirt

Don't just provide a "software solution" but provide it in a box, of your choice that you support. With the model your suggesting (BSD rather than Linux) you could do this entirely without having to go through any other provider which is awesome.

Still can't get over the M$ patents everything article, especially the bit where the authors suggests that M$ may be trying to patent web services. Next they'll try to patent XML .....

Z.

This is a really good argument for Open Source:

"Open-source applications...allow anyone to look at the source code. For major open-source applications, such as the Linux kernel, the Apache Web server, etc., dozens or hundreds of people will read the source code either to learn how it works, make modifications or look for mistakes," Trappe said. "Because the development process is also open, these independent reviewers can report the defects they find and even suggest appropriate fixes."

"Unfortunately, this process takes too long for most commercial product development cycles,"

http://zdnet.com.com/2100-1104-985221.html