Slashdot Mirror

Same here. So you me and the two ACs make 4. Foolishly I've even just released an app which I use on mine. Just one problem though, and it's now evident that it'll never be fixed -- the WinRT broker infrastructure that's responsible for throttling or allowing apps to run beyond their energy usage quota (on the user's request) is busted, so it's impossible to write a truly reliable connected-standby TCP socket app for Win10 mobile build 15063, the final build we'll probably ever see.

"WebExtensions" are just glorified Greasemonkey scripts -- with a much larger API than GM scripts have available, but you're still playing around in a sandbox. These are mostly interesting for extending how a given website works, hence my comparison to GM.

With WebExtensions, you're not playing in a sandbox, you're playing with abstractions. Exposing internal XPCOM objects to extensions offered no abstraction - extensions were messing directly with browser internals, which was severely hampering our ability to improve fundamentals of the Gecko engine. We want to be able to make significant changes to Gecko without constantly worrying about breaking extensions. An abstraction layer is the way to do that.

"Extensions" have access to the browser chrome, and can thus add or change any functionality in the browser. This is the bit that is euphemistically (and wrongly) described as "deprecation of XUL and XPCOM" (even though XUL and XPCOM aren't actually a requirement for this type of extension...).

You're right, XUL and XPCOM are not required for that type of extension. WebExtension APIs are going to allow for modification of browser chrome. Mozilla is working with extension developers to make sure that the functionality that they require will be exposed to them in a way that is future-proof. We fully intend to allow for extensions such as tree-style tabs and Vimperator.

If you have useful ideas for new extension APIs, I encourage you to offer suggestions instead of FUD.

WebExtensions overview
WebExtensions roadmap
WebExtensions experiments

Some roads are ment to be used heavily and some can be, but aren't. Suddenly change that and it causes all sorts of change, like demands for more road works to accomodate the increased traffic, falling house prices because the neighbourhood isn't quiet any longer, and all that. So no, this is not a binary thing. "The road is there, therefore it carries a right to send highway levels of traffic over it, in fact a duty to carry as many cars as will fit" just doesn't fly. But it is what waze assumes.

I say this is a defect that need fixing. Waze caused it, perhaps unintentionally, but as a direct result of their technology, so the least they can do is have a stab about alleviating it.

Say hello to the Xbox One screen saver fiasco.

You forget the team lead: an older woman who doesn't understand the difference between click and double-click and has to do all her testing with one of those laptops that has a virtual scroll bar on the trackpad.

But enough about Marissa Mayer! (It woud explain the multiple years of negative feedback they've ignored regarding the UX revamps that started when she got on board...)

But in terms of prioritising titles for support, I'd like it if MS could reflect on those titles where there is the greatest historical interest in ensuring they remain playable on new hardware

Actually, MS is prioritizing based on user votes. See http://xbox.uservoice.com/foru... for a list of the highest voted games.

The only reason games might not be ported despite high votes is the publisher - some of the top ones are Activision and EA, who will probably not allow those games be played on Xbone because they'd rather sell you a new copy of the game.

Of course, just like the 360's backwards compatibility, the compatibility is limited because it's all about he experience - backwards compatibility will be crappy if the game runs crappy, so I'm pretty sure MS vets all the games to make sure they at least run decently. Unlike the early PS3s which had a PS2 built into them, the 360 and One at best use an emulator, if not recompiled. Though I'd be surprised if they recompiled - game code is almost always lost the moment the game's released.

Vote for it here.

If I were in charge of that, you'd have it like yesterday :)

On a more serious note, at this point I wouldn't put it into the "never gonna happen" bucket anymore, just based on all the things I've seen the company do in the past year that were in that bucket two years before. But either way, it will take a long time - bash (and any Unix shell, really) really expects a lot of Unixisms from the environment that it runs it. Basically, I don't think you can get a proper *sh without having a proper POSIX layer underneath. And all we have today is Cygwin, which is basically a giant hack.

On the other hand, command prompt is getting some long needed love in Win10, and hopefully beyond. And when they asked about what people want from that effort, the requests for things Unix ranked pretty high on the list. These guys have said that they'll pay close attention to feedback, so I hope they'll deliver on that promise.

If I were in charge of that, you'd have it like yesterday :)

On a more serious note, at this point I wouldn't put it into the "never gonna happen" bucket anymore, just based on all the things I've seen the company do in the past year that were in that bucket two years before. But either way, it will take a long time - bash (and any Unix shell, really) really expects a lot of Unixisms from the environment that it runs it. Basically, I don't think you can get a proper *sh without having a proper POSIX layer underneath. And all we have today is Cygwin, which is basically a giant hack.

On the other hand, command prompt is getting some long needed love in Win10, and hopefully beyond. And when they asked about what people want from that effort, the requests for things Unix ranked pretty high on the list. These guys have said that they'll pay close attention to feedback, so I hope they'll deliver on that promise.

If I were in charge of that, you'd have it like yesterday :)

On a more serious note, at this point I wouldn't put it into the "never gonna happen" bucket anymore, just based on all the things I've seen the company do in the past year that were in that bucket two years before. But either way, it will take a long time - bash (and any Unix shell, really) really expects a lot of Unixisms from the environment that it runs it. Basically, I don't think you can get a proper *sh without having a proper POSIX layer underneath. And all we have today is Cygwin, which is basically a giant hack.

On the other hand, command prompt is getting some long needed love in Win10, and hopefully beyond. And when they asked about what people want from that effort, the requests for things Unix ranked pretty high on the list. These guys have said that they'll pay close attention to feedback, so I hope they'll deliver on that promise.

A drastic change of architecture like that with barely compatibility layer surely pissed of a lot of people.

And that's just the users, not the developers.

Dumbass wannabe solution architect?

But hey, Marisa told us to be agile(tm), so we changed everything and got it all done in sprints(tm) and even though the unresolved issues with the Finance and Sports sites continue to pile up and the userbase continues to dwindle away, at least we made a formerly functional UI use lots of AJAX, so even if it doesn't work for any of the users, we devs get to put all that Javascripty stuff on our resumes when we're looking for our next jobs. Thanks for the free career boost, Marisa! Remember, the only value your company actually has is the Alibaba IPO!

Then they change perfectly Gnome 2 with a half cell phone gnome 3/shell! Now office 365/office 2013 is all FREAKING WHITE IN ALL CAPS where I get a migraine looking at it. Then they change Hotmail.com to all blinding with blue. ,.. now gmail is changing too.

You forgot to mention the ever so popular UI changes to Visual Studio 2012/13. An utter UI train-wreck from MS, far worse in implication than Metro, as it is serious productivity s/ware used by Windows developers. And they haven't even backed down from using all-CAPS in the menus.

I've never trusted Microsoft since they tried to drop classic VB6

Vote for VB6

However, if slashdot dies because of this, it won't be because of Dice. It will be because like the dinosaurs, we couldn't adapt.

Well, first, why should a user base be required to "adapt" to a broken system? If they functionality they want goes away, they'll go somewhere else. Try to learn from other mistakes or you will die. Just check out all the comments there, most are along the lines of "Well it's too little and too late, me and all my friends have gone elsewhere."

Hmmm... linky failed. Should have been this

But we are listening and incrementally improving the experience based on what users are telling us.

Yea, you lost me right there. It's not an "experience" - it's a tool. When you call it an experience, we know you're using it just like others:

• The "experience" is why Digg changed their site and lost most of their users (to Reddit AND slashdot)
• The "experience" is why Yahoo changed their Groups and refused to change it back and took 6 months to give their users an emphatic "no", even though most of them had already left.
• The "experience" is the reason HTC refuses to provide Android updates for last year's model phones, or even fix any of the bugs because they're embedded in the firmware.
• The "experience" is the excuse HP, Asus, and Lenovo use for loading gigabytes of resource-sucking crapware and nagware on their consumer computers.

I could go on with this list extensively, but know that your audience understands this kind of marketspeak and translate it immediately into "We follow this policy that we know you will hate because we think it will improve our revenue." Review the results of the examples above and you will see how poorly this typically works out.

VS2012 never required IE10 and can do Metro apps. But yea, they had to change the VS2013 setup right after RTM to turn the IE10 check into a warning and later release an update that has additional fixes

There's a similar though smaller revolt going on over the changes to Yahoo's Fantasy Football. The nasty thing about the Fantasy Football changes is that they didn't roll them out until two weeks before the start of the season, after lots of people had already paid as much as $250 to join pro leagues.

Yahoo went so far as to post an announcement to every league that they won't be going back to the original format (but they really appreciate your comments!).

Because there is no website. See here.

When I first started using Feedly it was just a shiny veneer over Google Reader and it kind of still is. Notice that without the extension there's no login on their site. They've assumed your feed information from Google Reader into their extension, but they haven't yet produced a version that they can host. They probably didn't have hardware that could handle their user base offline before Google announced Reader is shutting down. The way the extension currently works, you could probably host feedly.com from a laptop.

Visual Studio needs to allow development for all the platforms. People will love it.

Many don't. Visual Studio pulled a Windows 8-esque faux pas and they also refuse to about-face on it:

As you read this post, those of you that are fans of the Visual Studio 2010 icon collection may be asking, “what about the icons?” We currently have no plans to offer the old-style icons for Visual Studio 2012. The Visual Studio 2012 icon designs are directionally aligned with the style used across a broad range of new Microsoft products. We believe this design consistency is important, and we expect the icon styles to become familiar and comfortable over time.

...Yeah. Having said that, other than the UI in 2012, Visual Studio is a fantastic IDE.

So early adopters of a completely new UI still prefer the old UI they've been used to for decades. Wow.

Microsoft particularly, however, aren't having a good time with their recent UI changes. The last two Visual Studio versions have been met with complaints. VS 2010 because of UI slowness and difficultly to customise. The latest, VS 2012, "Metrofication" created an uproar. http://visualstudio.uservoice.com/forums/121579-visual-studio/filters/top

I'm amazed VS2012 was released looking like this. Was there really enough positive feedback to justify it? was there a lack of negative feedback?

There were loads of negative feedback once the beta with the new theme was out. E.g. this is still the highest-upvoted user feedback item on UserVoice for VS, and you can also go read the comments to the blog post that announced the changes. What you see in RTM is actually better than what was originally there.

I can assure you that there is also a lot of negative feedback internally, myself included. Several developers have actually posted links to those UV items on their personal blogs to drive votes.

Is the capital letters thing part of Office 2013 too, or is it just VS that's like this?

It's there in the public beta of Office 2013. Whether it sticks for release or not is not for me to say. I would hope they'd learn something from our experience.

It should be noted though that other aspects of their theme are much better, IMO. They also went all in for the flat look, but they didn't convert icons to nearly monochrome - they're flat, but they still use colors, and in more or less the same way they've always been used (e.g. folders are yellow, and save floppy is dark blue). So overall it looks better. Also, for them, the ALL CAPS are used for Ribbon titles, which is not quite the same as main menu - as Ribbon is normally always expanded, and rather heavyweight, those caps don't look overly heavy and imbalancing.

Is there any possibility we can have a poll on whether the person behind this vandalisation of the interface can be taken out the back of your offices and shot too?

The closest you can get to that is voting for those two tickets (colors and caps) on UserVoice. I can tell you that a lot of people are watching those closely and getting regular updates on their status.

VS 2012 is not a Metro app. It's a desktop app that tries to look like Office 2013, and doesn't even do that quite right. None of the things that you're complaining about is Metro-related, anyway.

Some kind people *ahem* have snucked in a registry key to let you turn all caps in menu off. But please also vote here.

You can get VS 2012 colors back (or roll your own theme) with this extension.

Unfortunately, there's no workaround for icons.

Not for much longer. Microsoft has already removed support for XP from the current version of Visual Studio, just like how they removed W2K support from VS2008.

If by "current version" you mean the most recent released one, that's VS 2010, which runs on XP and lets you target it as well. If you mean the upcoming VS 2012, then by itself it won't even run on Vista, much less XP - Win7 is required. On the other hand, while it can't target XP out of the box for C++, the outcry about that was strong enough that support was re-added, though it will now have to come as a separate download because the decision was reversed so late in the release cycle.

Sure, you can still compile for W2K with VS2005, but for those who want to be completely legal about it, just try buying VS2005 licenses now.

Boxed versions of VS 2005 are still available. More importantly, MSDN subscription includes legal downloads for all Visual Studio versions all the way back to 6.0.

For VS 2010, C++11 (then 0x) implementation was actually better than g++. It looks dated today because g++ has more frequent releases and it didn't take it long to catch up and overtake. Ditto for Clang.

For VS 2012, yeah, it's lagging behind, mostly because a lot of time was spent on C++/CX language extensions for Metro. They did update existing features to be spec conforming - e.g. lambdas were previously implementing draft semantics which have changed in the FCD, so those were updated to match the latter. It also adds a bunch of minor stuff like range-for and strongly typed enums, and a good chunk of new libraries (most notably atomics, threads and futures). Here are the details.

There is talk about doing a separate release later on specifically for VC++ to catch up with C++11 support. If you care about that, rather than just seeing it as an opportunity for some MS bashing, please add your vote here (I did).

One lone blogger does not an outcry make.

No, but 1500 people who voted against this on UserVoice do. As do all the blog comments.

TFS is not entirely correct, either. The real story explains people outrage better. Let me explain.

"The recent release of Visual Studio 2012 contained a UI element that few believed could make it into the final version — ALL-CAPS menus. After lots of user criticism and disbelief, Microsoft has moved swiftly to do something about it — by tweaking the typography.

In fact, the first release of VS 2012 that contained ALL CAPS in some form was the beta, but there it was used for tool windows (panes), not for top-level menu.Here is what that looked like - note "SOLUTION EXPLORER" etc, but regular menu. Users complained loudly about it back then. It wasn't the only sore point, but it was one that was by far the most "popular".

Then RC comes, and removes ALL CAPS from tool windows... only to put them into the main menu instead. Needless to say, the users were not amused when that happened after the clear feedback that caps are pointless and annoying. Now there's a new UserVoice submission for that, which quickly gets into top 10.

And after all that, the linked blog post comes out that explains that caps are actually just fine and will stay, despite "some people" not liking them!

At this point, it's not really about caps themselves. As you say, there is a reg key to change it, after all. It's more about how user feedback was handled (or ignored) along the road.

TFS is not entirely correct, either. The real story explains people outrage better. Let me explain.

"The recent release of Visual Studio 2012 contained a UI element that few believed could make it into the final version — ALL-CAPS menus. After lots of user criticism and disbelief, Microsoft has moved swiftly to do something about it — by tweaking the typography.

In fact, the first release of VS 2012 that contained ALL CAPS in some form was the beta, but there it was used for tool windows (panes), not for top-level menu.Here is what that looked like - note "SOLUTION EXPLORER" etc, but regular menu. Users complained loudly about it back then. It wasn't the only sore point, but it was one that was by far the most "popular".

Then RC comes, and removes ALL CAPS from tool windows... only to put them into the main menu instead. Needless to say, the users were not amused when that happened after the clear feedback that caps are pointless and annoying. Now there's a new UserVoice submission for that, which quickly gets into top 10.

And after all that, the linked blog post comes out that explains that caps are actually just fine and will stay, despite "some people" not liking them!

At this point, it's not really about caps themselves. As you say, there is a reg key to change it, after all. It's more about how user feedback was handled (or ignored) along the road.

Wait, what, they made the buttons in VS11 flat and greyscale?

http://blogs.msdn.com/b/visualstudio/archive/2012/02/23/introducing-the-new-developer-experience.aspx
http://visualstudio.uservoice.com/forums/121579-visual-studio/suggestions/2623017-add-some-color-to-visual-studio-11-beta
https://connect.microsoft.com/VisualStudio/feedback/details/726448/make-proposed-monochromatic-ide-changes-optional-in-vs11

Last time, you nearly had me convinced that Microsoft is too busy doing other things, and we should excuse them for their failure to implement C99 (or all of C++).

But have you SEEN what the Visual Studio team has been working on with 11? That god-forsaken UI is the ugliest thing I've seen since Motif. And at least with Motif you always knew where one button started and one button ended. ... If THAT is their excuse for not implementing C99, may they burn in hell. Or at least their project managers.

Well, it's not like there is a single monolithic "VS team". There are many different teams doing different things, and, most certainly, it's not the C++ team doing the new theme in VS11. In fact, even for the team that does (which would be VS Platform team, and more specifically Shell), most of the work is done by designers - there isn't much dev time spent there. And devs are not easily swappable, either - C++ team is, of course, mostly writing in C++, and they don't really care about much else; while theming work is 99% WPF/C#.

And yes, as far as the theme itself goes, I agree that it's ugly, and it's not like it's even "Metro" (which was the original excuse). Don't ask me how this got through dogfooding, either - I'm too bitter about that to be objective.

Well, at least the people responsible know about that by now... so it's not the end of that story yet.

Furthermore, some projects break when you try to import them from VS2010.

If a VS10 project fails to open in VS11, that's almost certainly a bug. If you have a LiveID or don't mind creating it, file it. If you'd rather not mess with LiveID, I can file it on your behalf.

The more votes this gets, the faster it'll be.

That said, while VC is not exactly on the frontlines here, I don't think there's any compiler claiming full conformance as of today. For myself, I'm quite content with having lambdas, decltype, auto and rvalue references, and the only thing that's really still sorely missing is variadic templates. Library-wise, VC11 is already mostly conforming except for initializer lists (which are kinda useless without compiler support, in any case).

This is the best place to ask for such things at the moment - unlike Connect, which requires LiveID, it lets you login with Google or Facebook accounts, so it's easier to gather votes.

If you have any specific feedback on missing features, or scenarios that are slower than you think they should be, please post it on http://visualstudio.uservoice.com/forums (or vote on existing items if they describe your problem) - this is used directly in product planning to prioritize work areas. Even if it's a "well duh" kind of thing that should be obvious - it probably is, and people responsible for it already know, but when enough users raise their voice to complain about something, it always sends a strong message about fixing it in a more timely manner.

FCC should require all ISP who use data caps to in include in their ads how many theoretical hours it would take to hit the cap at the max advertised broadband speed.

Got a better Idea; tell the FCC

http://fccdotgov.uservoice.com/forums/105541-fcc-gov-feedback.

And the number 1 feature request is to implement their own login system so people don't have to deal with OpenID:

http://stackoverflow.uservoice.com/pages/general/suggestions/16685

OpenID is simply poorly-implemented. It's only useful for sites that store absolutely no data whatsoever; if the site stores data, OpenID actually makes the situation worse by having multiple web servers hold the data hostage. Sadly, they haven't even slightly improved on the Microsoft Passport system they were inspired from.

Also see this posting from yesterday's article where I outline how OpenID is worse than traditional logins, and their "solution" for that problem actually makes things worse, not better:

http://tech.slashdot.org/comments.pl?sid=1011619&cid=25556389