What I find more disturbing is the influx of Microsoft friendly commenters. Articles on various MS products are frequent and receive a lot of praise.
Articles that point out how MS is crippling innovation through extortion of Linux users receive relatively little attention.
Google is an extremely open company (with respect to user data and user rights as well as contributions to open source software). However, most comments about Google products and contributions seem to take the most negative view possible. Case in point, today's article on Richard Stallman's comments on Android focused on how he was critical of Android, mostly ignoring that he said "Android is a major step towards an ethical, user-controlled, free-software portable phone..." Android is still the most open major platform out there. It's put Linux on millions of devices.
> Android is still the most open major platform out there. It's put Linux on millions of devices.
If you want to do crazy stuff on your iPhone you jailbreak it. If you want to do crazy stuff on your average Android you root it. How does using Linux alone make something 'more open'?
I also think that most metrics for being "open" are missing the big point about the cloud. Yeah, maybe Google's business code is written in FOSS tech and they give people a FOSS browser and one or two FOSS mobile OS to use their services. Where's the practical open-ness for the end user? It's still a cloud ecosystem, the one thing that is worse than a closed-source desktop ecosystem.</rant>
> Why desperate? This is a valid way of advertising a product.
If my boss were to come to me and tell me "let's draw a big blue arrow over out entire front-page" I would have just laughed in his face. Now I'm afraid it will be harder for me to do that.
Let me elaborate. Google.com it's the ultimate advertising resource in the Internet. Google+ is now being accused (i don't have an opinion, i don't have an account and I don't have the real numbers) of stagnation and Google just:
1) opens Google+ to everyone
2) makes this arrow
It is a cohesive move: You put an arrow because it's open and it's open because of the arrow.
But, giving the timming, one could think that someone at the C level at Google agreed to use 'the nuclear option'.
If that doesn't work for Google+, Google is out of options to push Google+ for good (it could run tv ads. but it won't, and it would be expensive and not really a good way to promote a social network).
Also, other people could find it confusing and reject the idea just because it's too "right on their face".
But, it could work super! That would be bold (because of the risks named above) and great for Google & Google+
I don't think Google necessarily needs to get all their users in one day of viral goodness like most tiny startups do. They have pleny of time and money to burn through; look at how long Microsoft has lasted. Google can gradually integrate their other products with Google+ so that you don't even know you're using +. One day you'll want to share something, and you'll already be on a social network, and it will Just Work. (This is especially relevant with smartphones. Think about further refinements of the Camera app's "instant upload".)
Anyway, I see slow growth for Google+. Most people don't want "social networking", they want to share photos with their kids or share funny jokes with their coworkers. Google+ will gradually integrate itself with services people use for this sort of thing, and will replace it as the "Tumblr feed" model replaces email, IM, and SMS.
Don't let the names scare you, they refer to the licenses not the quality of the code. Normally these would be detected and installed automatically, but there's a bug that prevents mms:// content from being autodetected.
In fact it seems that it's impossible to install by clicking on the link
"One moment, please, while the current Silverlight installation status is determined...
Microsoft Silverlight may not be supported on your computer's hardware or operating system."
I'm sure this was meant as a joke, but, the downmods still make me sad. I used VS for some classes in school and at my first job and it made me a worse programmer. It took years to undo the damage. In my opinion, as a developer, and in my experience in working with a lot of other developers, terminal + vim + build tools (gcc/javac/python, etc) is far superior compared to VS.
I'd be curious as to what 'damage' you needed to have undone. I've written in just about every IDE available (including vim) and quite frankly my opinion is that, well, tools can make things easier but they won't make anyone a good programmer.
The debugger is a big one. I had a terrible habit of not thinking through the code but just running through the debugger until I got it working.
Not thinking about how code looks (especially to other people) since I could just resize the window to whatever.
A lot of the build complexities were hidden, which was fine as long as everything worked, but, as soon as something small broke, I spent days reading documentation and uselessly searching MSDN and the web trying to get the project to build (so, not understanding my build environment is something else I had to fix).
There are lots more (source control, collaboration, etc). I know that switching my tools and my development environment made me a better programmer.
I do most of my development in Eclipse or VS, because frankly my time is worth more than screwing around with Vim for more than basic text editing. Funnily enough, I only break out the debugger when something's really nasty--maybe once a week, for half an hour or so. I certainly don't practice debugger-driven development, and I don't know anyone who does. Personal problem, not a tooling problem.
"Not thinking about how the code looks"--personal and cultural problem, not that of the tools; whether I'm in Vim or Coda or VS2010, I run a guideline at 100 characters (because 80 character lines is dumb in the age of the smallest dev monitor being 1280 pixels wide) and wrap to it. Never heard a complaint from anyone I work with.
Build complexity--here, you have something of a point, but MSBuild or Maven or whatever isn't appreciably different than tooling with makefiles (except that MSBuild probably has better documentation and there's some semblance of a standard way to do things).
Source control? I use Hg and Git, seamlessly, from within VS2010 or Eclipse. Not a problem at all. Collaboration? I've used VSTS for it before, but I just hang a Chrome window in my second monitor and keep Redmine (at home--Bugzilla, at work) and an IRC channel open. None of this is appreciably easier in the Land of the Holy Terminal--it's just different.
I've used both heavily IDE-driven stuff and the stone knives and bearskins that vi-land suggests, and my practices do not fundamental change from one to the other--because I understand what I am doing and how and why to do it. Changing your environment doesn't help you become a better programmer unless you understand why you did what you did and have developed the discipline to not do it regardless of tooling. Your problem existed between keyboard and chair, and it doesn't sound like you've fixed it--just hid it away.
My understanding of the comment is that the commenter makes the point that VS is almost too comfortable (sorry if I'm mis-reading this), and that comfort is actually a hindrance to learning things.
I see it where I teach, the students that come in are most comfortable with windows + VS and they are productive in this environment. However, after just one course in C programming using vim, the students have a much greater appreciation of the process of producing software. They gain an understanding of compiling and linking, Makefiles and automated builds. (I'm not sure all of them like vim + make, but they see where VS is automating stuff for them and I think this is a valuable thing to learn).
Do you need this understanding to be a productive programmer? Of course not and indeed learning it is a bit of a timesink compared to getting your features finished.
Eventually like most things it is horses for courses (it doesn't make sense to me to develop C# code using vim as you have the worlds best(tm) IDE to help you), but personally I feel that IDEs have become too much of a crutch, they help so much that you can be productive without knowing what you are doing and, this maybe elitist, I feel this harms the industry a little.
Good programmers will always be good programmers, bad programmers can hide with an IDE, they have to learn what they are doing when dropped into vim + make (or other lo-fi tools of choice).
I have coworkers who use Eclipse (for Java). It's a CPU hog; often it locks up entirely, crashes are infrequent but, happen. I've tried it; I know I'm much faster with vim. For me, my time is worth more than putting up with Eclipse.
I use 80 characters per line because if I am at a larger monitor, I can have 2 or 3 split windows side by side. If I'm at a smaller terminal, everything's still readable. I find narrower code is more readable anyway. Most developers I admire write like that. In any case, this is a major flame war topic; if you don't like it, I'm glad that your system works for you.
As far as build systems, I'll take autotools and its learning curve and pain over VS any day. I can fix it when it breaks.
What are you trying to accomplish by saying that my problem is not the tools but that I'm bad at programming? Why is it that because you think you can write decent code in VS, that means that I didn't learn anything new when I switched from it? I know I certainly did; in fact, if I could give myself any advice back when I started any sort of formal programming, it would be: get off VS and Windows immediately! I would be three years ahead of where I am now.
I'm not saying you're bad at programming, I'm saying you're blaming the tools for your lack of discipline. Plenty of good programmers I know are undisciplined. But you can learn, and it's not the tools' fault that you didn't do so before. Hell, I'm a VS2010/Eclipse driver but I write MSBuild scripts by hand and I spent time the other day debugging the make/ant monstrosity. Limiting yourself because it's the only way to make yourself learn what you need to learn may work, but why not just learn it?
(And re: Eclipse--yeah, it's a bit of a hog, but hardware is cheap. Developer time isn't. And the features it provides have no equivalent in vim et al.--just look in the Source dropdown menu for a number of significant productivity enhancers. Those are some of the more minor ones, even; Open Type probably saves me half an hour a day in the monster of a codebase I have to work in. Eclipse is by no means perfect, but it's geared toward reducing boilerplate and stepping on the annoyances involved in Getting Things Done.)
Most of my peers in school used the same "compile, debug" approach as well. Most of them are still doing it and I can't see any serious software company ever hiring them (there is no shortage of positions for mediocre programmers though, so, they're doing just fine). I don't think this symptom is unique to me and my lack of discipline. As to why I didn't just "learn" to use VS or "learn" to program better? It wasn't for the lack of trying. MSDN was worthless, written documentation (yeah, it was that far back) was worthless and all of the code that other people wrote in VS that I had to interact with was absolutely terrible. It wasn't until I switched my environment until that I got any better. I got more from a man page in an hour than I did from a week of reading MSDN.
As far as Eclipse; I've worked with large codebases and I was faster with vim and knew my APIs better than any of my colleagues who used Eclipse. I heard them complain about Eclipse freezing pretty much daily (as well as doing something wrong when it came to interacting with perforce, often in a devastating way). I know that I Got more things Done than any of them.
That's an interesting take. You do have to have tools that adapt to your personal style, and it sounds like you felt that you needed to push yourself into different habits, which is commendable.
I've found that many programmers who come from a Java background over to .NET have a bad habit of writing their code and deploying it without debugging it even once. It's so much trouble to debug for them, they just wait until it blows up and takes down the server. Or their error handling is just throwing a stack trace to a screen instead of actually handling the error.
Heh, yeah, that approach sounds pretty terrible to deal with. What I do is extensive unit tests and useful logging. Unit tests catch most of the stupid stuff and plenty of the sophisticated stuff (and are a useful thing to have for a lot of other reasons anyway). Logging lets me track down what slips through the cracks. Most of the bugs that resulted in serous production issues for me were a result of rather complex interactions between complex components, not something a debugger would have helped anyone with.
That said, I have no doubt that there are a plenty of people who are amazing at writing code in VS and are much better developers than I am.
I remember attending a presentation by Elon Musk on SpaceX a few years back (late 2007, I believe). At that point, SpaceX's future was very uncertain with their two launches failing.
I recall him describing the Proton as the most reliable vehicle ever engineered by humans.
I think that SpaceX is great, but, is there really a gap to fill, in this case? Where does this gap begin?
the gap is absence of competition. Monopoly situation is bad for everybody, including the monopolist itself who degrades into the state of not giving a damn f&ck about any-/everything.
? The gap is in time--the time between now and when the next resupply mission will be authorized to fly. In the case of Progress, it is uncertain when that will be.
