We publish all our desktop apps using ClickOnce, which is built into Visual Studio. It is so much easier than the installer. Easier for the developer to set up and publish new versions of the app. Easier for the user because we configure it so that apps always check for updates when the app starts and automatically applies the update if one is available. We use this for desktop apps distributed to people who work at our office and for an app we developed that is run by external people across the country.
I worked as an install developer for over four years for a large shop--yes,I actually spent 8+ hrs/day doing nothing but install dev. Because I now work for a small company w/many small projects, I made heavy use of the VS Setup template because it was quick and easy. When I ran into the sun-setting of Setup templates, I did a systematic search for a 3rd-party installer. While I did find several products that would have worked, the best value for me was Advanced Installer by Caphyon. The Pro version is just $299, it's easy to use, it works, and the vendor publishes frequent updates with fixes and enhancements. They also have bigger dollar versions in case I ever need it, and a free version. I was able to do things like consolidate 32- and 64-bit installs into a single installer, easily install two windows services (I wrote in VB.NET) in a single installer, specify pre- and post-build actions just like in VS build, and launch my install builds from the cmd line using VS post-build actions. It's not the Holy Grail, but worth every penny. And no, I don't work for Caphyon. Hope this helps.
I have only one copy of Visual Studio left on my computers. Once the things I have to finish up are done, this copy will also be going to East Hyperspace. They wanted to make certain that we all follow their great ideas, instead they assured that I forget about them as quickly as possible.
Even if they realize what went wrong at some point, I still don't trust them anymore. They will do it all over again at the next opportunity. Just stop giving your money to Mickeysoft and look for someone who does not treat you like a fool.
The Expression Blend tooling required to work with WPF/Silverligh has been droped.
All open support tickets seem to have been closed as 'don't care'.
And yes I know Blend has been integrated into VS, but its not feature complete and its not robust.
So moving forward you have no real support and no real migration path.
The tooling in VS will gradually get worse and worse (just like it did when they 'continued to support' C++/MFC).
Agree entirely. Microsoft's sudden about-turns have really damaged my trust in them.
We used the WPF platform and I really liked it. Sure it had problems, but it had a number of positives.
Now, if I start writing a new desktop app tomorrow, for the first time ever I have no idea what language and framework I should use for the longest possible 'active' development support life-cycle. Crazy.
Given the pain that Silverlight is causing us I'm glad it's going, that way I don't have to talk people out of using it.
But yeah, there's other sh*t MS is doing that is pissing me off. Forcing standards which only they use for one. They don't have any clear direction on authentication and authorization between their tool sets (OAuth, Claims etc...), the disappearance of the installation project template.
Then there's the love which C# is getting from the company as well at the moment. Frankly this just leaves me in a place where I prefer using products which have not originated from MS
Eagles may soar, but weasels don't get sucked into jet engines
Couldn't agree more. The issue I have with this decision from Microsoft isn't that Silverlight was the best thing that've happened since sliced bread, but rather that I find it disturbing that there are no good alternatives (as of today).
Our main application is based on a Windows client that they normally run, but with a "light version" using Silverlight for when they're our of office or otherwise unable to use the normal client. For this purpose I still haven't found a nice and yet somewhat easy and efficient way to do this, except for... Silverlight. Flash won't cut it. Java applets (yeah right) and ActiveX components are just dead and buried.
So, what to do?
Thank you MS.. For yet another kick in the nuts, I'll be the first to leave for LAMP or similar when it's mature enough to completely wipe our present MS based platforms from the face of the earth.
Any web application is accessible from a connected device, even on a Mac (Mind blown!)
I could boilerplate something that replicated your grids functionality in a matter of hours and be assured that it would not only work both cross-browser but cross-device from mobile to tablet to desktop.
"[A] dozen different languages and technologies"...
I hate to break it to you but Silverlight is not the language of the web. It never was and never will be. If it was you might have an argument but it's not. The three aforementioned languages are and they are never going away.
I suggest you pick up your toys, put on your grown up trousers and accept that you are going to have to join the rest of the world. Or you can keep stacking cans in your bunker, up to you.
"He took everything personally, including our royalties!"
David St.Hubbins, Spinal Tap about Ian Faith, their ex-manager
From what I understand MS is abandoning Silverlight because they no longer need to compete with Adobe Flash as an addon for the majority of hand held devices that are currently in use because they don't allow addons. However the MS developer community and many large companies have embraced Silverlight as the UI of choice. Silverlight is widely used and is a fantastic product. I think MS is making a knee-jerk reaction and needs to realize that there is no reason to give up on Silverlight.
PRO TIP FOR MICROSOFT: Vision isn't following other's lead. Vision is often recognizing that you've the winning hand right NOW and simply going with it. Retards.
Interesting that you should put it that way. I couldn't have said it better. Fairly recently (after playing with Microsoft's latest "stuff") I realized that it simply isn't a direction that I want to go. I'm hunkering down with my system and tools that I've paid for exactly as they are and plan to stick with them for the long haul. I develop to the desktop and the web. Unless something major changes (I mean a complete platform re-design of the web or something) these tools will last for a long, long time. I'm done chasing the upgrade curve. It's going to be a LONG time before a Winforms or Web application won't run on just about any machine I care to deliver it to.
Since making that decision I have been highly productive and creative again instead of wasting time evaluating, installing and uninstalling all these new tools.
I'm not against "progress" but some days I view the changes taking place not as progress but mainly a re-design of everything so they can just sell more of 'em.
I think what I find astonishing about the whole thing is that Apple is making a huge killing with an incredibly closed environment. The 'write once, run everywhere' mantra sounds good on the surface but I just don't see where end users care about that sort of thing. Most people just want what they have to work and that's about it. Despite the conventional wisdom, the fact that Apple stuff doesn't work everywhere has resulted in... *drum roll please* ... more sales for Apple.
(Setting aside the fact that write once, run everywhere has always been a bit of a lie. No serious business application can be pared down from a 22" monitor to a 4" smart phone without quite a bit of extra work)
If I were Microsoft I'd make sure that I had a standards compliant HTML5 development environment and I'd slavishly support it on every form factor. No doubt that is required. In fact, I'd produce a lite version of Visual Studio that was little more than a HTML5/CSS text editor and give it away for free.
The difference is that I'd still develop Silverlight and make sure that it is better than HTML 5 in every way that that it runs better on Win Phone, Win Tablets, and Win PCs. While the W3C was still busy trying to find their own asses with their hands tied behind their backs I'd be rolling out fantastic stuff in Silverlight.
Microsoft has the money and power to do both of these things very well.
You know, the thing (to me) that seems to hamstring us developers is that we get convinced by the marketing forces that be that we have to be "everything to everybody". The idea of "write once run anywhere" sounds good but in actual practice it's just not likely to happen.
Where in the rule book (if there is one) does it say that by being a developer you have to be able to produce something that will run on absolutely everything, everywhere, huh? For the longest time I was caught up in the idea only relatively recently discarding it as impractical. You might say that I've finally "smelled the coffee" on this one. (It only took me 36 years in the industry to really "get it").
I used to think like that, what a letdown was such a decision, since then I'm full time focused on WP and W8 so I could reuse skills, and fortunately I got a job where I can just do that.
Now I have a different perspective because I believe the very reasons Microsoft droped Silverlight were:
1. They were not winning against HTML as web platform standard
2. For XAML/C# to win as a UI platform everywhere they have to aim for native apps first, starting from their own flagship product: Windows
3. The HTML strategy in W8 is to attract web developers and eventually they will end up developing more native apps than web apps.
I think at the end of the day HTML can't beat native in any platform, think Phonegap for example, so I think Microsoft is not joining the web standards completely, they are supporting native apps and while doing so they are empowering their own platform and others so web use in mobile is coming to an end subtly. I think that's more clever than being stubborn about Silverlight.
And by the way, I believe they are supporting Xamarin heavily because of this, supporting native we all win
The people saying HTML is the future, well, they are already living in the past in my opinion
All I know is that I spend more time configuring .Net and less time writing code these days. I agree that the standard Desktop and Web apps will be around for a long time, at least until some new MS OS comes out and makes the Federal and State governments add some mass quantities of patches/misery. Nice, their trying something new, but why not just improve on the good, existing, and since we worked like hell to make them working/reliable .Net IDE tools. I program from Main Frames to desktops/webs, to include Cobol,C#,vb.net,old vb6; last thing I want to do is learn some new code for some spiffy flashy Krappy app. True, everytime MS does come up with something good, they either quit supporting it or make a miserable attempt to improve it. Maybe its the high turnover up there in Redmond;
Just got back from lunch (yes, alcohol was involved; but not to the extent that I can no longer stand).
I've been doing a few sods and odds for a friend's company and we where disgusting the bids he's had for a new web site. I know slightly less than Richard the Retard concerning web design but I was able to offer one wee nugget - if the solution is driven by the technology rather than the technology driving the solution then it is the wrong solution. That killed five from seven tenders and I am on a retainer to sort out the others.
I may well be able to pay part of my gin bill this month.
Reality is an illusion caused by a lack of alcohol
the solution is driven by the technology rather than the technology driving the solution then it is the wrong solution
If the solution isn't driven by the technology de jour then you're going to have no end of trouble when the poor marketing people have to actually focus on the product rather than the technology. They may even have to use a thesaurus.
Symantecaly yes, but the question is which comes first? If you've decided on the technology without looking at the problem then you are failing. The technology is a tool in providing the the solution rather than technology decides what the solution will be.
As an example, too often I have seen poorly designed web-apps where a simple desktop solution would be a better fit; and vica-versa. I strongly submit to the concept of deciding the technology after understanding the required solution; maybe I'm just old fashioned.
Reality is an illusion caused by a lack of alcohol