A significant amount of my time is spent in Visual Studio, Microsoft Office and Firefox (IE beta 2 is too slow and buggy, Chrome not feature rich enough).
Can you explain how this will improve my using all the above software?
That is a very accurate assessment.Yggdrasil said:vesuvius said:*snip*
If so, it seems like a good evolution. I've always felt that WF had great potential for abstracting common patterns that many data-driven and process-driven applications share, but integrating it into a project required basically building your project around it. It's a framework rather than a library. If it can be made simpler and easier to integrate, and to use only as much of it as you need, it could be great.
Oslo is actually quite boring for most developers anyway, that would far rather be working with Flash/Silverlight, html or a general purpose programming language (s).Yggdrasil said:Dexter said:*snip*
When I build a data-driven business application, most of my effort lies in making it simple to develop, simple to understand and easy to change and extend when the inevitable change requests come, not squeeze every possible CPU cycle. This is what M/Quadrant are for, building modular applications easily. If there's one language it WON'T replace (well, two) it's C/C++, which are languages mostly used for processor-intensive applications, snappy GUI applications, low-memory scenarios and other situations which are entirely irrelevant for M's goals.
Oslo is based around Web Services and Windows Workflow both now termed Workflow Services (it is a domain specific language - and not general purpose). All this stuff is based on .NET so M will be an abstraction to solve a specific Oslo problem, but I think .NET is already flexible enough anyway, and I don't know how this "M" will increase, promote or tackle concurrency.
Invariably folk get "ants in their pants" about a new programming language, but I'm yet to see posts by a 9'er targeting workflow foundation, consequently I'm rather nonchalant about M.
Well Done!ScottWelker said:Before allowing this thread to die I thought I’d provide an update. All developers were granted local administrator privileges.
This is more than I sought; the “Least-Privileged Account” idea seems more prudent. Aside from advising management, I played little role in the resolution. It turns out that corporate policy is for developers to have local administrator privileges. It was our site’s Infrastructure team that was not following standards.
It seems worth noting that, thanks to you contributing 9ers, I caught my “solutionizing” mistake. I refocused my efforts and argument on the Why instead of the How:
1) Change happens, especially in technology fields like software development. Developers must be able to research new tools, technologies, and techniques - even ones that may go nowhere and that would never pass a formal review/approval. Much is often learned during the journey, even when/if it does not lead you to the desired destination.
2) “If the only tool you have is a hammer, it is tempting to treat everything as if it were a nail.” Often the only difference between timely success and complete failure is applying just the right tool. Developers must be able to bring various, sometimes new and different, tools to bear.
3) Software craftsmen must have many tools in their toolbox, including their tried and true, trustworthy tools.
**Note: For better or for worse, our shop is hyper customer pleasing. We are forced to rapidly adapt to our clients needs – and technologies. It’s how our business is presently structured. We could argue the merits of this approach but it doesn’t change today’s reality.
Thanks again everyone for your help!
I certainly thought "that was that!".
I guess from the purists point of view HTML is the way to go. I did use webforms etc. in ASP.NET 2.0 a while back and whenever a control came out be it a menu control, grid control or whatever I found loads of people moaning about the extraneous markup generated. That and view-state ending up in CSS adapters and so forth for the menu.Ion Todirel said:I would use ASP.NET MVC, besides it's a good opportunity to learn a new technology and possibly Silverlight in some places, depends on what type of web site tough, what I'm very sure I wont use are ASP.NET Web Forms
I found the whole experience very frustrating, especially coding for all different browsers. Progress can be so slow when you have to fix the same problem in different ways for the same code in different browsers, luckily I won't have this problem. Is the listview control in ASP.NET the best datacontrol to use, or does the repeater, datalist or other grid controls still have their uses?
I'm just about to complete a business (enterprise sounds bigger and better doesn't it) application that is n-tier. A definite desire is to consume the application on the web as well and I really needed to try and determine which approach to use.Maddus Mattus said:I would go for MVC, in combination with EntLib. David Hayden has some excellent posts how to integrate Unity into MVC.
I've written classic ASP.Net websites, websites with the Web Client Software Factory and recently I've done a pet project with MVC. I like MVC best, because of it's simplicity.
The gap between the HTML made by the designer and the HTML that your website generates, is smallest with MVC. It's more manageble then classic ASP.Net. You are missing some of the features, event validation, viewstate, etc. but that's no biggie.
The code gets a lot cleaner. No more need to know exactly in what phase of the ASP.Net Page Lifecycle you need to do what, I've always found that one of my largest pains.
The only drawback of MVC is the lack of Ajax behaviours/controls and the ISAPI filter thingie in IIS 6.0 (hosters do not like putting wild card ISAPI filters in).
Oh, and HTML designer support tends to get a bit freaky when you place a lot of code tags in.
Silverlight is attractive because it is cut down "WPF" which it would be nice to learn (all those routed events and so forth). Only problem is that I need the web part up and running quick, and learning something new is not the best approach. StackOverflow was done with MVC in a relatively short time, but you had people that knew what they were doing. The web application will not be for public consumption, but for management to check on how operations are performing (it's a 24 hour thing), and their partners to have limited access to the work-flow information as well (basically it is retail, so they need to know which purchase orders have been processed, and when to expect them etc.)
If you had the choice to create a new website, would you use the MVC framework and pure HTML approach, or would you choose Silverlight?
Assume that there are no financial constraints, but the project must be completed within 6 months, with a working shell ready in at least 12 weeks.