Web Development

Ask something
Discussion Locked

Don't JUMP to .net

0 helpful
26 replies

by petergatzios

Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

I can't find a simple, straightforward answer as to why I should move away from the technology I'm using and go to .NET. This site http://msdn.microsoft.com/vjsharp/jump/default.aspx let's me know HOW but not WHY. Well, they do give one reason, so I could take advantage of xml. (Which I can do anyway). But here's what I'm using my technology to design http://www.iokio.com/myff/gm_used_2_door/ .
Technology will come, and technology will go. Why should I consider this .net framework as having any more staying power than let's say "the video phone"?
Peter
p.s. btw, to me it seems like PASSPORT is losing ground. Maybe chalk this one up as well?

26 replies

by e1-2wdq-nnqb-snwg

If you're using flash, it might not be worthwhile. Same for Dreamweaver if the web-site is very content-intensive and does not have much of a "business application" aspect to it.

ASP.NET is a "run, don't walk" change from ASP, and many other text/code-intensive methods of dynamic web page generation. It excels at business applications, especially complex ones that might have been done a few years ago as a Windows (non-browser) application.

Windows stuff is still the way to go for very rich user interfaces, such as drag and drop, but forms and such can usually be done with ASP.NET, including treeviews and tabs. Third-party controls for those items as well as menus make development fairly easy.

People who find ASP.NET hard to use effectively are usually public content website designers, while former windows application developers find it a very productive platform (not as easy as windows where desktop caching is available, but still very productive).

by petergatzios

From reading this and the other posts there seems to be agreement on investing along with Microsoft when it comes to future proofing either your application or your employment. You just have to stay current and this is the latest flavour of the day. I remember when a client tried to move their mainframe business logic to a browser back in '98. The integration with the authentication engine (RACF)and JAVA almost killed the whole project because it took almost ten seconds for authentication to occur. What a nightmare but Java was the future for everyone that wanted a more open standard than IE specific integration. (Could the last person at Netscape please turn off the lights).
So, I suppose now it might get into a religious discussion of open standards, integration and my personal favorite "best in class development environment". To which I've learned that this last one depends entirely on the spacer (the person between the chair and the screen). :)
So, it looks like I'm going to look into this .net more seriously. Anyone have a "Coles Notes" to .net?

by cq_west

The biggest advantage that I've discovered for implementing and developing against the .NET platform is the ability to quickly develop functional and robust applications. .NET is a framework that you can use to bring rapid development of systems together.

If you are using VB, the new VB.NET encompasses some very powerful new OO (Object Oriented) programming. No longer is VB the inferior language.

The biggest advantage that I have been able to find is in the development of enterprise class applications. The .NET framework takes care of alot of the plumbing so you no longer have to worry about it -- however, if you want to build the plumbing yourself, there is still plenty of flexibility to do that.

You mention XML as one of the reasons mentioned. To say that XML is the only reason encompasses a whole library of functionality. The new ADO.NET is XML enabled, allowing you to model data in memory with constraints and relationships enforced. The dataset offers some very powerful functionality when combined with a data adapter. A complete database update or loading of table data can be performed by the execution of a single command.

Overall, the framework is focused on allowing a developer to deliver fully functioning applications in less time than previous development systems.

by petergatzios

Let me declare up front that I'm not a big fan of MS. As a business I think they are a great example to emulate and aspire to. As a technology company they really have struggled to "play with the big boys" and only recently developed applications that can handle millions of transactions per second. I know they also get hammered for security but it's important. And I imagine if any other company had the same amount of attention their software would be proven just as vulnerable. But when you're number one you can't be as good as the next guy, you need to be better, much much better. And MS has proven that they are not. Just ask Bill every time he does a live demo.
If you're invested in MS skills than keep going. Nobody has the "best" tools so why not stick with what you know. As for .net masking the complexity of the lower level plumbing, well...you then have to accept that the folks at MS have done a skillful and efficient job. Maybe it's snobbish to say this, but MS is still a desktop company, a much better desktop compamy than ten years ago but it's not a teraflops player. (64 bit software anyone?) But maybe that's not a bad thing given that 80% of the world is only interested in browser based interfaces and email. :)

by cq_west

I appreciate that you indicated you are not a fan of MS up front. It helps me to better understand your perspective.

One of the big pushes from Microsoft lately has been the development of smart client applications. The idea is to have the richness of a desktop application and still the connectivity of a browser based system.

Also, Microsoft is working to build 64-bit versions of their software so that developers can utilize the features and speed of this new architecture.

The problem with much of Microsoft software is that it is almost too easy to use. This allows for people who have not developed solid design patterns and practices to develop software that "works". While the software works, it may not make use of the best patterns and therefore is riddled with bugs.

Some other operating systems and development platforms are almost seen as elitest by many developers. Compare the number of Linux Developers and then the developers that use Visual Studio (or some Microsoft Platform) and you'll see a stark difference. Many of the Linux developers have special training that allows them to leverage the O/S and development platform properly, while anyone that can write a "Hello World" application in Visual Studio calls themselves a developer.

I have much of my training the the Microsoft World, but I'm finding that anyone who has developed good design strategies can implement solutions that work on a host of platforms. (Just look at the gaming industry).

by itsfantastic

One of your concerns seems to be that the technology may disappear as quickly as it arrived - while possible it's highly unlikely, since Microsoft are basing so many of their own products on the technology and including it as core parts of operating systems and other server infrastructure.

As for why should you move away - there is no definitive answer to the question. If the technology you're using now is allowing you to do what you want then why move to something else? If you need to do things you haven't done before then that may be the time to move.

It definately is a very powerful technology and is worthwhile investing time and money into learning and using, but it has its pitfalls like everything else.

If you're looking at doing anything with web services - either building or consuming - then .Net is streets ahead of anything else at the moment...but they're catching up.

Finally - IMHO you should use whatever is the best technology for the job, and not get hung up on using only one technology versus another

by xquery

almost any kind of job in web-based application related with xml-xslt transformations (get xml request -> get params -> make some calls to web-services -> put xml and some params to related xslt -> apply interaction logic layer -> apply design layer -> apply device dependency layer -> perform output as xml (xhtml, wap) or pdf and so on

in conjunction with MS SQL 2005 (with robust support of XML) ASP .Net 2.0 is great tool for orchestrating web-services into web-pages (just look for example at XsltCompiledTransform)

don`t compare present - better compare future, at least near future :)

by petergatzios

So, web services are what's important and not the way we get there? I would agree with that if the XML standard(s) ever get sorted. In the end the customer doesn't care what the back end is or where the source was compiled. As long as they can quickly get what they want they will keep coming back. The frustration is with the people who have to serve up the result set. Which reminds me, what ever happened to JINI (another promising standard that never got out of diaper stage). Hopefully XML will make it to the teenage years.

by petergatzios

Cameron, sound advice. As for web services, I am using them now and I am sure that a customer will come along that will want to integrate to whatever they use so I just have to ensure my aplication calls can be executed. Ah, the beauty of standards, so many to choose from.

by petergatzios

I think you mean "like sticking with DOS". (You see that would be a software reference).Oh wait, MS still uses that as a building block don't they? Never mind.
Thanks for the comment anyway.

by manuel_morales

"Jumping" to .NET is just a small "step" and not a jump, at least for those who have been using other technologies than Microsoft ones.

At least for me (an old Java professional) moving to .NET languages was.....very easy.

I think that people who come from the "old" Microsoft programming model (no OOP, like VB, ASP) simply can`t see the whole thing that .NET is. IT's not a set of languages, it's not a way for using XML, it's simply a change about the way you do programming . It is a revolutionary change to Microsoft "philosophy" and put this company in the way that since years ago ANOTHERS companies are running on. All the "new" features available in .NET were in Java since almost...7 or 8 years ago, so for Java oldies moving to .NET is..simply. To me, .NET is the official MS way of saying "well we were wrong, ok.". the CLR, .NET Framerowk, Web Serrvices, Secure code and all those "advantages" are not new features, compare it with other tehcnologies availble on SUN and...they are conceptually a copy (MS would you like to talk about pattents and intelectual property ?).

If you NEED to do it, well ok. If you are just thinking about it, first think about "why?", "what technical advantange it offers to me?".

by careed

Noticing that the Microsoft link that you referenced is dealing with J#, if you're happy with Java, then go with it.

The switch to .NET is the same as with switching to any other programming paradigm. You go with the flow and hope that you don't get caught in the whirlpool.

Take Java as an example. Sun, while still buying up other smaller companies, may be a prime candidate for purchase by someone else. If that someone else is like IBM or HP, the software side of Sun may be spun off to a new company that will have to compete directly with Microsoft and Borland (plus all of the many Linux/Unix software providers out there). On the other hand, if it's bought by Microsoft, Borland, or even Novell, the software may be incorporated to that company's own offerings. Because Java never made it to true open standards status, anyone acquiring Sun could even squash Java like a bug.

Microsoft, while not the greatest of companies, can probably provide one some confidence that it will still be around within the next ten years. Granted, .NET may have morphed into something else, but, then again, what language hasn't. I'm sure that you would find more AcuCobol users today vs. users of the original version of mainframe COBOL; more FORTRAN 90 users than FORTRAN IV; more Visual Basic users than QuickBasic, etc....

I truly believe that one should never reject a programming paradigm based on how one feels about its manufacturer/distributor or about the way it works. Personally, I wouldn't want to program in COBOL; but if I have to, so be it.

Overall, programming should be fun. Whether you're using .NET, Java, or FORTRAN (yeah, I still like FORTRAN), if you like it, keep doing it. If you happen to get a paycheck along way, so much the better.

by petergatzios

You're right, personal feelings aside there are a number of development environments that are purpose built and do a great job within their defined parameters. And I would certainly use them if I needed to pay the mortgage.
And nice reference to Cobol, I was reminded of that stuff called Turing. Man that was a lot of HB lead we used to get "Hello world" to show up on a green screen.

by absolutezero

No matter what flavor of development languages / technologies you use, they all do the same thing.

The real difference between programming languages is not in what they do well, but in the hard things they make easy.

Although, I can program VB and C# .net, I have not embraced the technology 100% and still tend to favor C++ and Java. The reason behind my decision is one of flexibility.

Embracing the .NET platform will make your applications totally dependent on Microsoft and require a large overhaul to cross platform deploy. A prime example is some of the companies I have developed web enabled applications for require all artifacts to be built using Java / javascript since their web platforms run on Sun Solaris or HP/UX.

Back to my original statement, you have to use the product you feel most comfortable with, enjoy the most, and gives you the flexibility and/or power you need to complete the task at hand.

by davemori

If it does not give you a quantified benefit or a high perceived value, and if your existing technology is fine, the cost of jumping to any technology probably exceeds the benefit.

IT is not a bottomless well of funding.

Microsoft does have plans to no longer require Passport authentication - but it is a ways in the future with Longhorn/Vista.

Frameworks EVOLVE. They take time to mature to something that has a compelling developer and solution set.

by petergatzios

Thanks for the comment I have to admit I'm more about getting to 90% with what I have than getting to 99% with something new. BTW, where does "laziness to learn something new" fit into my CBA? :)

by dr dij

He might go broke if you personnally don't shell over a bundle for the latest versions, as quick as he can shovel them out.

So we can change the paradigm, from languages that used to work pretty much unchanged for decades to languages that have a half life of 5 years.

For your fellow workers who need jobs recoding perfectly fine company software that is now in an obsolete language.

by petergatzios

I agree, but I'm with Bill. A guy that actually labeled his version of Office so there was no mistake as to the life expectancy. '98, '00, '03, '06 anyone?

by djameson

I love .net, the biggest advantage is when it compiles, it compiles to a common machine language. which means with 20 Mb or so of common librairies installed on the machine virtually any application in .net will run. webapps thick, thin clients. you write in c# c c++ perl java j++ etc it all compiles the same.

by petergatzios

Not this post, the one I'm replying to. Good point and one for .NET. However, I'm weary of MS tools given their history of adding code specific to MS targets and not staying within a defined standard, even a loosely defined one.
And I would like to return to lean coding where the answer isn't just beef up the host client or count on relatively fast internet access speeds. Which reminds me of a great story I heard a long time ago from a consulting friend. This guy was in Redmond walking along a hallway when the marketing guy asks the windows guy what the spec should be on the side of the box for the new release of windows? The tech guy says "take what the last recommendation was and double it". Nice.

by synthetic

After reading all the replies, I don't feel I can add much to your questions, cost benefit, expanded options, do you trust MS, etc.... It has been covered. I did want to say that I am impressed, having answered back an provided thoughtful opinions, insights, and simple thank yous to all repliers is pretty cool. I also wanted to take a quick moment to commend your ikio feature finder. Great design work from the link. Seems a very useful tool!
Johnathan

by sherman.meeds1

VS.NET represents a powerful expansion of development capability, with much, much reduced time to develop PLUS an much improved iteration of OOP principles. Once experienced, you have a methodology that provides quick development capability. Once you develop a good set of function libraries and user controls, it becomes simple and quick to develop even very complex applications.

by ianr41

I think the issue of .NET or J2EE (or any of the other program platforms) both a business and engineering decision. Engineers can be bloody minded purists...without much business sense. On the other hand...business people can be bloody minded egoists, without any engineering sense.

But, the .NET versus J2EE is starting to gather evidence that .NET is becoming both a mature and well engineered development platform...and is making a whole lot of business sense - in that projects arrive more often on time, on cost - and .NET performance and cost comparisons with J2EE for small to mid-range commercial projects are very much in favour of .NET. I think the jury is still out on massive projects. For a number of reasons, not least of all, the superiority of high-end Unix iron. But even here, a number major projects, e.g. Netbanking, have been successfully re-developed in .NET, running on Oracle and clustered Intel architectures.

I hate the notion that .NET is silver bullet. It is still maturing. There is still a need for strong .NET frameworks. But it is moving forward - and developers are putting forward convincing reasons why projects should be done in .NET.

Neither J2EE, php, ColdFusion, are going to die anytime soon. But .NET is a worthy platform and C# is very efficient, and of course, the whole caboodle is XML native. And while XML is a verbose neighbour...I do think it is here to stay (well, 5 years in this game is a lifetime).

by petergatzios

Thanks for your comments, you mentioned re-development of some projects. This reminds me of the discussions I've had with CIO's that have determined their preference for Microsoft/Intel/Oracle vs Unix-Linux/non-Intel/DB2 packages. And most of the decisions were then based on experience rather than benchmarks. More often than not, labour was an issue and getting MS skills were more abundant and less expensive.
I don't like to fight a bigger opponent, and MS is as big as they come, but I have to admit I'm going to take a couple more jabs before I throw in the towel. Besides, I've always chosen the under dog. Much more exciting.

by lchapman

It's a double-edged sword. On the one hand, you want to keep current, because as one replier states, technology does move forward, so you can't remain static (for long) - you're either moving with the current, or you're sliding back.
On the other hand, and I feel this one too, MS has changed VB, ADO and ASP so much in .Net that it's almost like starting over. I would like to have seen them add the new features without destroying the legacy features. So much doesn't work in VB.net that did is VB6. I could go from VB 4 to 5 to 6, and still carry most of my skill & experience with me. That mostly stopped with .Net.
I believe the bottom line is that if you're going to be in the Technology business, you'd better be open and acceptable to change, even if you don't like the change. MS's changes WILL go forward, like it or not, so best view it as progress, accept the inevitable, grab a (another!) book on it, and move with it.
Today there are a few more job openings for a .net developer, and a few less for a vb6 developer. Next year this time, you can place money on the fact that the numbers will have grown by ten fold (or more). And yes, at this point, I still prefer VB6. Ask me next year this time, hmmm??

Good coding!
Larry

This discussion is closed to new replies.