It's really *not* "very, very hard" to do using only Subversion. Yes, you need to get your head round "I want the changes from this location and this time copied over to this *other* location" but once you've done so, it's really not terribly hard to do exactly what you were saying.
Reading the red book gives a good feeling of the whole "time+location" business - I don't see the problem. If you want to use svnmerge to make life easier for you, that's fine, good for you - but I really think you're overplaying how hard it is with just svn.
I regularly used to merge revisions in the way you suggested. (I now work with TFS, which is much nastier when it comes to merging. Blech.)
I set up a small (two line) shell script so I wouldn't have to manually do the -rx:x+1 each time, but it was absolutely fine. In what way is it "very hard" to do with Subversion for you?
You're confused about the "public API" part of this. As far as I understand it (given what Jamie's written) it uses one public API for the commercial SKUs, and a *different* public API (related to user controls) for the express SKU. Both APIs are public though.
I'd be willing to bet a good quantity of imaginary dollars that the number of Mobile Phones, Wiis and other gadgets using Opera is far smaller than the number of computers using Firefox. I wouldn't like to guess at the install base in terms of numbers of installations, but where the numbers are much more in Firefox's favour (I would hazard a guess) is in terms of the amount of browsing done with each installation.
I own a Wii, and I've got Opera on it. Both my wife and I have laptops with Firefox on. Does that mean that our view of the web is on a 1/3 Opera, 2/3 Firefox basis? Not at all - we very rarely use the Wii for browsing, but we browse with Firefox all the time.
How many people spend more browsing time on a mobile phone than they do on a computer? It may increase over time, but I suspect that traffic going through Firefox is *much* heavier than that going through Opera, for that reason.
I've got a blog entry on this. Some of my comments were off base because I didn't know about some of the features of VS2005 at the time, but the ones I really would like fixed: 1) Open Resource (Ctrl-Shift-R in Eclipse) 2) Open Type (Ctrl-Shift-T in Eclipse) 3) Organise imports (using directives in C#; get rid of ones I don't use, sort the ones I do, add any necessary) 4) Make the overload tooltip show more than one ruddy entry at a time!
Oh, and compile-incrementally-on-save-so-quickly-you-don't -notice is rather handy too.
No, if it's identical it just means that you have misunderstood the various articles you've posted links to, and they're not talking about issues which mean the November 27th code isn't ready for consumers - they're talking about issues which meant that Vista wasn't ready to meet its *original* release date.
The delay the articles talk about wasn't for nothing - "the delay" means that Vista is better than it would have been if it had been released early in the year.
The delay between the consumer release and the business release was always going to happen, and is due to things like production, giving PC manufacturers more time to include Vista in their "out of the box" installations etc.
1) 1digit.blogspot.com: (December 2nd) "Microsoft said today that consumer versions of its next-generation version of Windows would not hit the market until January 2007, dashing plans that it would be on PCs for sale during the 2006 end-of-year holiday season."
From that same page: "And just what is this so-called RTM? It's the major milestone where we can confidently say that Microsoft is done with Windows Vista and will be handing it off to our industry partners: PC makers, ISVs and IHVs." What does "Microsoft is done with Windows Vista" mean to you? To me, it means their part of the code is complete. If hardware manufacturers improve some drivers between now and January 30th, that's fine, great, and I'll update appropriately - but that's entirely different to the picture you paint. Of course, you appear to believe you know more about what Microsoft release than they do...
2) www.cio.com entry: Well, that's from July 11th, well before the November release, so is irrelevant: There's nothing in there which suggests the code in the November release is different to the code in the January release.
3) Washington Post: Again, nothing in there suggests to me that there's any difference between the November and January releases.
And yes, it states that hardware support and software compatibility support was a factor in the fact that Vista has released later than was originally intended. That's *not* the same as saying that there are any compatibility issues which produce a delay between the November release and the January release.
You seem to want to treat the word "delay" as if it always means "delay after shipping to businesses but before shipping to consumers" whereas in almost every case shown so far it actually means "delay between when we originally intended to release Vista and when we actually shipped it". Can you see the difference?
*All* releases were delayed, compared with when they had originally intended to be released.
That is a very different matter from whether the delay between the release being available on MSDN and the release being available in a box is due to more code changes being required or production matters (boxes, discs etc).
In my view, there was *always* bound to be a difference between when download availability and "in the box" availability.
But hey, we'll see - as I said, we can compare the disk images after the "in the box" release.
I didnt announce my product wasnt yet ready for consumer release becuase of issues that I think are just as (if not more) pertinent to commercial customer...
Could you post an article/reference to where MS said this after they'd released it (as RTM) on MSDN? You see, November is significantly after March (which is when the pages you link to were written) - so just because they said in March that there was more testing to do doesn't mean there was more to do after they released it to MSDN.
I very much doubt that the version shipped to consumers in January will be *any* different to the version on MSDN now.
The reason it takes longer to ship to consumers than to businesses is that discs need to be printed, packaging needs to be created, etc - whereas businesses can download from MSDN and install directly. Do you think MS is still actually cutting code to be part of the January release?
Of course, it'll be easy to tell after it's released in January - take a disc image of the version you get in the box and compare it to a disc image available from MSDN now.
Yes, Vista was delayed. Yes, that changed the date it got into customer's hands - but that *also* changed the date it was available on MSDN. If the final version had been available in a box by now, the downloadable version from MSDN would have been available earlier too.
In what way am I a troll? Does the fact that I have a different opinion on Microsoft than you do make me a troll? My post was simply correcting a common misconception: that all MVPs love Microsoft.
Note that there is a difference between not loving Microsoft and out-and-out hating them. I don't love Microsoft. I don't hate them. I dislike some of their business practices (certainly historically; I feel they're improving, although it would be very good to see the rate of change increase). I dislike their public attitude towards Open Source, particularly Steve Ballmer's. Then again, I dislike the attitude that many Open Source folk take towards Microsoft, too.
Note that my post did *not* defend Microsoft anywhere - and many of Roblimo's criticisms seem pretty reasonable to me. One interesting point to consider is that being respected by Microsoft to a certain extent means that criticisms from the mouths of MVPs tend to be taken somewhat more seriously (which is not the same as saying we always get our own way, by any stretch of the imagination). You'll find plenty of MVPs who are vocal critics of various aspects of Microsoft, whether in terms of technology, product evolution or business practices.
I post a lot on the C# newsgroup. After doing that for a while, I was nominated, and I've been renewed every year since. I don't think I'd say I actively sought it out, but given the amount I was posting when I was first awarded, it would probably have looked slightly odd if I hadn't at least been considered.
You are always awarded based on previous community contributions, not future ones - so to stay as a C# for *this* year, all I have to do is stay out of trouble (not violate the NDA etc). Of course, if I don't do anything for the community, I probably won't be reawarded next year.
Some of the benefits are probably under NDA, but there are obvious ones: freebies, better contact with MS employees, and the MVP summit in Seattle.
"There are people who love Microsoft. The company has an active Most Valuable Professional (MVP) program that encourages outside volunteers to help other users."
Now, this doesn't specifically say that MVPs all love Microsoft, but I think that's the conclusion most people would draw from the above statements. As an MVP (C#) I'd just like to say that MVPs don't all love Microsoft. I'm more positive about MS than I used to be, partly as a result of meeting some great and really smart employees, partly due to some good technologies coming out of Redmond (along with not so great ones, certainly) - and no doubt freebies have a certain amount of influence.
However, this doesn't make me a Microsoft shill, and it doesn't mean I dislike non-MS software where appropriate (for instance, I prefer Eclipse to Visual Studio, even though I prefer C# to Java). In the MVP community there's plenty of irritation with certain bits of Microsoft. MVPs are often valued within the community because they're not shills, and won't always say things are rosy. I'm not saying we're completely unbiased - MS treats us very nicely, and we'd have to be inhuman not to be swayed at all by that - but that's a long way from the implication of the quote above. I've certainly never had any pressure put on me to be "nicer" about MS in newsgroup/blog posts.
If you *ever* touch the network, you really shouldn't do so in the UI thread. A good example of the problem is Firefox, which will hang completely if some Javascript makes a network request. That's a really nasty situation to be in.
The difference in user experience between a properly written multi-threaded app and a single-threaded app which doesn't take great care not to make blocking calls is vast, IMO. That doesn't depend on whether or not you've got multiple processors - it's just a case of blocking calls not preventing the UI from working. Yes, most of the time many apps will be waiting for user input - but the computer doesn't mind that, whereas users really *do* mind it when they can't even resize or close a window because a server they're talking to hasn't responded yet.
How, exactly? If you use non-blocking IO, then in many frameworks you'll end up being called back in a different thread - which leads to all the normal threading problems. Writing code using asynchronous IO is far harder than writing it using blocking calls - and harder, IMO, than writing a multi-threaded app using blocking calls.
Now, in certain situations an application really, really shouldn't use any blocking calls - but for a normal GUI, I'd far rather do long-running tasks in a separate thread and marshall calls back to the UI thread than try to make everything asynchronous.
If the docs are stored as XML files, then I suspect Monad would make things a *lot* more robust - you could load the XML files up *as XML documents* then use your favourite navigation technology (XPath, XQuery, whatever) to find the nodes you wanted. The only bits which need to use plain text are the first and last ones, and they can be done easily using the.NET framework.
Now, maybe it isn't obvious exactly how you do all those bits - but I suspect if you gave someone bash and only told them as much as you know about Monad, they wouldn't find it obvious how you do the above task either.
I had this yesterday, too. I believe it's due to Windows Live changes. Try following the "log in with a different email address" link (or whatever it is - you'll know the one I mean), then use the same address again, and things may well be okay. I had to do the above and then start from the download page again, but I got in that time.
While I suspect the grandparent's estimate is high, you've misinterpreted it by two orders of magnitude - it wasn't $0.02 per click, it was 0.02 *cents* per click. Still a lot just for doing a search though.
No. 45% of Americans who responded to the survey finding out why they didn't have broadband say that it's too expensive. In other words, it's ignoring the 61% (according to the article) who already have broadband.
The way it's written in the summary is like saying that 70% of people don't drive because they don't have a licence, even though clearly more than 30% of people *do* have a licence. (Okay, those numbers were completely made up - I hope you see my point though.)
Do you have any examples of the 1MB apps you're talking about? Most AJAX usage I've seen has been really small.
As for doing things quickly - I totally agree, which is why I infinitely prefer the Google maps way of doing things to the "old" way where you were basically in the "North by a square, north east by a square, only show me 3x3 squares at a time" trap. Did you really find that more usable than Google Maps?
As others have said, Visual Studio is significantly behind Eclipse. I believe it's also behind IDEA, although as I haven't used IDEA in anger I wouldn't like to speculate.
I blogged about just what I miss from Eclipse when I'm working in Visual Studio. Visual Studio 2005 has made up some ground, but it's still miles behind IMO.
One thing to note: Eclipse has had the "press enter while writing a string" functionality for a little while. I think it came in the 3.1 series. That's one of the problems of comparing these IDEs - I know that Eclipse develops pretty quickly, and I'd imagine the other two do too. Keeping up with what each can do would leave little time left for coding!
Slashdot Mirror
It's really *not* "very, very hard" to do using only Subversion. Yes, you need to get your head round "I want the changes from this location and this time copied over to this *other* location" but once you've done so, it's really not terribly hard to do exactly what you were saying.
Reading the red book gives a good feeling of the whole "time+location" business - I don't see the problem. If you want to use svnmerge to make life easier for you, that's fine, good for you - but I really think you're overplaying how hard it is with just svn.
Jon
I regularly used to merge revisions in the way you suggested. (I now work with TFS, which is much nastier when it comes to merging. Blech.)
I set up a small (two line) shell script so I wouldn't have to manually do the -rx:x+1 each time, but it was absolutely fine. In what way is it "very hard" to do with Subversion for you?
Jon
You're confused about the "public API" part of this. As far as I understand it (given what Jamie's written) it uses one public API for the commercial SKUs, and a *different* public API (related to user controls) for the express SKU. Both APIs are public though.
I own a Wii, and I've got Opera on it. Both my wife and I have laptops with Firefox on. Does that mean that our view of the web is on a 1/3 Opera, 2/3 Firefox basis? Not at all - we very rarely use the Wii for browsing, but we browse with Firefox all the time.
How many people spend more browsing time on a mobile phone than they do on a computer? It may increase over time, but I suspect that traffic going through Firefox is *much* heavier than that going through Opera, for that reason.
Jon
Who says you have to be a consumer-level user to want to install a consumer-level version of Vista?
It would be nice to be able to test whether an app works on all versions of Vista without having to have them all on physical boxes.
I've got a blog entry on this. Some of my comments were off base because I didn't know about some of the features of VS2005 at the time, but the ones I really would like fixed:
t -notice is rather handy too.
/ 22/79631.aspx
1) Open Resource (Ctrl-Shift-R in Eclipse)
2) Open Type (Ctrl-Shift-T in Eclipse)
3) Organise imports (using directives in C#; get rid of ones I don't use, sort the ones I do, add any necessary)
4) Make the overload tooltip show more than one ruddy entry at a time!
Oh, and compile-incrementally-on-save-so-quickly-you-don'
Here's the blog entry: http://msmvps.com/blogs/jon.skeet/archive/2005/12
I haven't tried Orcas yet (hope to do so with the March CTP really soon) - perhaps some of it will have been implemented there...
Jon
I'm sure you're right about solar power being a big part of the future - but maybe not in Scotland ;)
No, if it's identical it just means that you have misunderstood the various articles you've posted links to, and they're not talking about issues which mean the November 27th code isn't ready for consumers - they're talking about issues which meant that Vista wasn't ready to meet its *original* release date.
The delay the articles talk about wasn't for nothing - "the delay" means that Vista is better than it would have been if it had been released early in the year.
The delay between the consumer release and the business release was always going to happen, and is due to things like production, giving PC manufacturers more time to include Vista in their "out of the box" installations etc.
Jon
Okay, let's look at each of those links in turn:
c hive/2006/11/08/windows-vista-releases-to-manufact uring.aspx
1) 1digit.blogspot.com: (December 2nd) "Microsoft said today that consumer versions of its next-generation version of Windows would not hit the market until January 2007, dashing plans that it would be on PCs for sale during the 2006 end-of-year holiday season."
Actually, MS had announced the January date much earlier than that: http://windowsvistablog.com/blogs/windowsvista/ar
From that same page: "And just what is this so-called RTM? It's the major milestone where we can confidently say that Microsoft is done with Windows Vista and will be handing it off to our industry partners: PC makers, ISVs and IHVs." What does "Microsoft is done with Windows Vista" mean to you? To me, it means their part of the code is complete. If hardware manufacturers improve some drivers between now and January 30th, that's fine, great, and I'll update appropriately - but that's entirely different to the picture you paint. Of course, you appear to believe you know more about what Microsoft release than they do...
2) www.cio.com entry: Well, that's from July 11th, well before the November release, so is irrelevant: There's nothing in there which suggests the code in the November release is different to the code in the January release.
3) Washington Post: Again, nothing in there suggests to me that there's any difference between the November and January releases.
And yes, it states that hardware support and software compatibility support was a factor in the fact that Vista has released later than was originally intended. That's *not* the same as saying that there are any compatibility issues which produce a delay between the November release and the January release.
You seem to want to treat the word "delay" as if it always means "delay after shipping to businesses but before shipping to consumers" whereas in almost every case shown so far it actually means "delay between when we originally intended to release Vista and when we actually shipped it". Can you see the difference?
*All* releases were delayed, compared with when they had originally intended to be released.
That is a very different matter from whether the delay between the release being available on MSDN and the release being available in a box is due to more code changes being required or production matters (boxes, discs etc).
In my view, there was *always* bound to be a difference between when download availability and "in the box" availability.
But hey, we'll see - as I said, we can compare the disk images after the "in the box" release.
I didnt announce my product wasnt yet ready for consumer release becuase of issues that I think are just as (if not more) pertinent to commercial customer...
Could you post an article/reference to where MS said this after they'd released it (as RTM) on MSDN? You see, November is significantly after March (which is when the pages you link to were written) - so just because they said in March that there was more testing to do doesn't mean there was more to do after they released it to MSDN.
I very much doubt that the version shipped to consumers in January will be *any* different to the version on MSDN now.
The reason it takes longer to ship to consumers than to businesses is that discs need to be printed, packaging needs to be created, etc - whereas businesses can download from MSDN and install directly. Do you think MS is still actually cutting code to be part of the January release?
Of course, it'll be easy to tell after it's released in January - take a disc image of the version you get in the box and compare it to a disc image available from MSDN now.
Yes, Vista was delayed. Yes, that changed the date it got into customer's hands - but that *also* changed the date it was available on MSDN. If the final version had been available in a box by now, the downloadable version from MSDN would have been available earlier too.
Jon
In what way am I a troll? Does the fact that I have a different opinion on Microsoft than you do make me a troll? My post was simply correcting a common misconception: that all MVPs love Microsoft.
Note that there is a difference between not loving Microsoft and out-and-out hating them. I don't love Microsoft. I don't hate them. I dislike some of their business practices (certainly historically; I feel they're improving, although it would be very good to see the rate of change increase). I dislike their public attitude towards Open Source, particularly Steve Ballmer's. Then again, I dislike the attitude that many Open Source folk take towards Microsoft, too.
Note that my post did *not* defend Microsoft anywhere - and many of Roblimo's criticisms seem pretty reasonable to me. One interesting point to consider is that being respected by Microsoft to a certain extent means that criticisms from the mouths of MVPs tend to be taken somewhat more seriously (which is not the same as saying we always get our own way, by any stretch of the imagination). You'll find plenty of MVPs who are vocal critics of various aspects of Microsoft, whether in terms of technology, product evolution or business practices.
I post a lot on the C# newsgroup. After doing that for a while, I was nominated, and I've been renewed every year since. I don't think I'd say I actively sought it out, but given the amount I was posting when I was first awarded, it would probably have looked slightly odd if I hadn't at least been considered.
You are always awarded based on previous community contributions, not future ones - so to stay as a C# for *this* year, all I have to do is stay out of trouble (not violate the NDA etc). Of course, if I don't do anything for the community, I probably won't be reawarded next year.
Some of the benefits are probably under NDA, but there are obvious ones: freebies, better contact with MS employees, and the MVP summit in Seattle.
"There are people who love Microsoft. The company has an active Most Valuable Professional (MVP) program that encourages outside volunteers to help other users."
Now, this doesn't specifically say that MVPs all love Microsoft, but I think that's the conclusion most people would draw from the above statements. As an MVP (C#) I'd just like to say that MVPs don't all love Microsoft. I'm more positive about MS than I used to be, partly as a result of meeting some great and really smart employees, partly due to some good technologies coming out of Redmond (along with not so great ones, certainly) - and no doubt freebies have a certain amount of influence.
However, this doesn't make me a Microsoft shill, and it doesn't mean I dislike non-MS software where appropriate (for instance, I prefer Eclipse to Visual Studio, even though I prefer C# to Java). In the MVP community there's plenty of irritation with certain bits of Microsoft. MVPs are often valued within the community because they're not shills, and won't always say things are rosy. I'm not saying we're completely unbiased - MS treats us very nicely, and we'd have to be inhuman not to be swayed at all by that - but that's a long way from the implication of the quote above. I've certainly never had any pressure put on me to be "nicer" about MS in newsgroup/blog posts.
Just thought I'd try to clarify things a bit.
If you *ever* touch the network, you really shouldn't do so in the UI thread. A good example of the problem is Firefox, which will hang completely if some Javascript makes a network request. That's a really nasty situation to be in.
The difference in user experience between a properly written multi-threaded app and a single-threaded app which doesn't take great care not to make blocking calls is vast, IMO. That doesn't depend on whether or not you've got multiple processors - it's just a case of blocking calls not preventing the UI from working. Yes, most of the time many apps will be waiting for user input - but the computer doesn't mind that, whereas users really *do* mind it when they can't even resize or close a window because a server they're talking to hasn't responded yet.
Jon
How, exactly? If you use non-blocking IO, then in many frameworks you'll end up being called back in a different thread - which leads to all the normal threading problems. Writing code using asynchronous IO is far harder than writing it using blocking calls - and harder, IMO, than writing a multi-threaded app using blocking calls.
Now, in certain situations an application really, really shouldn't use any blocking calls - but for a normal GUI, I'd far rather do long-running tasks in a separate thread and marshall calls back to the UI thread than try to make everything asynchronous.
Jon
Only if you don't mind your GUI completely freezing if you make a blocking call of any description.
If the docs are stored as XML files, then I suspect Monad would make things a *lot* more robust - you could load the XML files up *as XML documents* then use your favourite navigation technology (XPath, XQuery, whatever) to find the nodes you wanted. The only bits which need to use plain text are the first and last ones, and they can be done easily using the .NET framework.
Now, maybe it isn't obvious exactly how you do all those bits - but I suspect if you gave someone bash and only told them as much as you know about Monad, they wouldn't find it obvious how you do the above task either.
Jon
I had this yesterday, too. I believe it's due to Windows Live changes. Try following the "log in with a different email address" link (or whatever it is - you'll know the one I mean), then use the same address again, and things may well be okay. I had to do the above and then start from the download page again, but I got in that time.
Jon
While I suspect the grandparent's estimate is high, you've misinterpreted it by two orders of magnitude - it wasn't $0.02 per click, it was 0.02 *cents* per click. Still a lot just for doing a search though.
45% of Americans say it's simply too expensive.
No. 45% of Americans who responded to the survey finding out why they didn't have broadband say that it's too expensive. In other words, it's ignoring the 61% (according to the article) who already have broadband.
The way it's written in the summary is like saying that 70% of people don't drive because they don't have a licence, even though clearly more than 30% of people *do* have a licence. (Okay, those numbers were completely made up - I hope you see my point though.)
Jon
Do you have any examples of the 1MB apps you're talking about? Most AJAX usage I've seen has been really small.
As for doing things quickly - I totally agree, which is why I infinitely prefer the Google maps way of doing things to the "old" way where you were basically in the "North by a square, north east by a square, only show me 3x3 squares at a time" trap. Did you really find that more usable than Google Maps?
Jon
...is create good IDEs.
As others have said, Visual Studio is significantly behind Eclipse. I believe it's also behind IDEA, although as I haven't used IDEA in anger I wouldn't like to speculate.
I blogged about just what I miss from Eclipse when I'm working in Visual Studio. Visual Studio 2005 has made up some ground, but it's still miles behind IMO.
Jon
One thing to note: Eclipse has had the "press enter while writing a string" functionality for a little while. I think it came in the 3.1 series. That's one of the problems of comparing these IDEs - I know that Eclipse develops pretty quickly, and I'd imagine the other two do too. Keeping up with what each can do would leave little time left for coding!