"VS is not only great"
I suppose this is something reasonable people can disagree about. I don't consider VS to be "great".
Microsoft has agreed to acquire development platform GitHub in a deal worth $7.5bn, sending developers scurrying for cover. The Office-maker is chuffed to call itself the most active organisation on GitHub, claiming more than two million commits made to projects. The Seattle software slinger has its origins in development …
"I don't consider VS to be 'great'."
*ESPECIALLY* versions after 2010. 2010 is the _LAST_ VS I will _EVAR_ use.
1. The newer ones have a 2D FLATSO look, jammed down your throat (or elsewhere)
2. The newer ones are oriented around 'The Metro' and UWP, from what I can tell
3. I thought the BEST IDE was DevStudio '98 where you could do EVERYTHING without lifting your hands of the keyboard, including the class wizard and dialog editor. It's been DOWNHILL since "DevStudio .Net" in the early 2000's
The _ONLY_ feature that I like in VS that's NOT in anything else is "virtual space". That's it.
Autocomplete gets in my way [I shut it off]. Background code scanning (for class info, etc.) is irritating, but I haven't had to shut it off (yet). The VB-ness of the UI, particularly 'property' displays for things like dialog box entries, are CUMBERSOME and require TOO MUCH "mousie, clickie, mousie, clickie" and moving my hand between keyboard and mouse, finding home row again, etc.. PAIN IN THE ASS.
VS is an example of what _MIGHT_ happen to GitHub, too. Wait until it becomes all 2D FLATSO and "touch friendly" and CIRCLE-JERKS you around when you try to find something in the help system (like MSDN has become). Too many pages with too little information, and ginormous 2D FLATSO "buttons" on them to go elsewhere, ending up back at themselves after a few mis-directions. Circle-Jerking.
> VS: *ESPECIALLY* versions after 2010. 2010 is the _LAST_ VS I will _EVAR_ use.
Exactly. Visual Studio VC++ 6 ('98), VS 2005 and VS 2010 were great.
Everything else incl VS 2003/2004, 2012, 2013, 2014,... and VScode (aka Atom.io fork) were worse and shit and contained spyware.
Xcode, IntelliJ, and even Eclipse and NetBeans are more in favor these days than the buggy spying shit from M$$$.
What is it with you people. How can you work in IT for any period of time and yet still be so frightened of change? "Oh, I'm not familiar with that, they changed to all capitals for the menu headings, I simply can't use it!"
Maybe think about changing career to something a bit more sedate?
Unluckily, there's been a time when too many people blabbed around that after Eclipse IDEs would have been just commodities and there were no need to develop others.
The result was just a stall in IDE development, and their dreadful state is well under our eyes.
There was also the vi crowd who never understood some class of applications - especially those with a GUI - are far better handled using an IDE.
I am writing my own. it's taking too long.
In the mean time I use pluma and csh/bash to do things. Sometimes I even edit with ee or nano [especially Makefile where you need hard tabs].
newer pluma has a built-in that lets you trim the extra characters off of the end of a line, too. Unfotunately not yet in devuan (last I checked), but FreeBSD's ports system has had this change for around 2 years.
who needs devstudio? I don't, not for open source on non-windows systems at any rate (or in cygwin).
I don't know your needs, and so I don't know what would be a better development environment for you. I'm not saying that VS is bad. I'm saying that it's not really accurate to portray it as if it were substantially better than its competitors -- but how true that is depends on your use case.
Why not? The list of alternatives supplied here is shorter than in my original question. Good to hear that Bob's is still in development but what are the other IDEs that VS isn't substantially better than? Just saying "it's not that much better" is no help if you don't add "than" and give alternatives.
'That's because you know of a better IDE!! Go on, enlighten us!
'(Last time i asked this the best responses i got were "eclipse isn't too bad" and "I'm writing my own".)'
Midnight Command isn't too bad as an IDE, and I'm writing my own, version of Midnight Commander, which is currently on GitHub, but likely gonna move it elsewhere soonish.
I've ben using BitBucket for years (before there was GitHub), initially because it was the service I was recommended and it used Mercurial, which I prefer but YMMV, but later on because I compared the T&C's and business models: BitBucket always had a revenue stream which gave it focus and less of a need to pimp user data; GitHub's business model was always hoping to find a way to monetise user data. And now it looks like they've succeeded. You can do a back of an envelope calculation as to how much MS thinks your code and programming metadata are worth for when they fold the service in LinkedIn and try and sell metrics to employers.
GitHub used to have the richer ecosystem for things like CI but these are now generally available for any (git) repo and BitBucket has continued to add its own features such as Pipelines.
"We're shutting down our repository"
I'm going to play 'wait and see' for now. My business does not require github (or anything like it) for success. However, it's probably an enhancement. A lot of open source code projects will remain, at least for a while. I'll wait and see where they end up, since I'd probably need to do pull requests on THAT system anyway if I wanted to submit anything.
Then I'll probably leave my stuff 'as-is' (for now) on github, and make them a "clone" of the _REAL_ repository "elsewhere" [sort of what I had done with stuff I had on sourceforge], continue to update it, and so on.
THEN, if things become 'really heinous', I'll dump my 'clone' repos [since they won't be "the official site"] and say 'why' on the MAIN site.
In any case, no need to 'rage-quit'. Not yet. Wait and see. And watch my ass.
"In any case, no need to 'rage-quit'"
I didn't "rage quit". I quit because I prefer to avoid Microsoft anything to the greatest degree possible. Microsoft is a horrific company.
Right now, I'm keeping my own Git repo on my own servers. Whether I'll move to a third party Git host eventually, I don't know. I'll keep an eye on the situation and if one appears that looks reasonable, I'll move to it.
...although reading they will be staying independent it looks like Chris will become a technical fellow at MS so will have to work for them.
This happened with WinInternals. When they got bought out by Microsoft it made Mark Russinovich and Bryce Cogswell rich men (I assume). However, both had to become Technical Fellows at MS first for a few years as part of the deal. Appears Mark enjoyed it and is still there. But Bryce, the day the clause was up, retired. Whether that's due to not enjoying working at MS or just deciding he'd rather just enjoy life and the money he now has only he knows.
> Sysinternals IS a good set of tools
"Sysinternals WAS a good set of tools" - fixed that for ya.
Unfortunately, Mark isn't working on them anymore. Rare updates and regressions all over the place. I am staying with older known to be good versions of his tools. The same with Windows, trusted Windows 7 keeps running great until the 2025 and beyond on modern 2018 hardware. And I laugh at all the Win10 sheeps that rely on spyware these days. Sad.
Does it sling ads at people? Charge money for businesses to keep their code there?
Looks like Microsoft has found yet another way to throw away piles of cash on failed acquisitions. I'm sure glad Apple isn't so dumb as to be splashing out $7.5 billion on a company that's probably worth a tenth of that at best.
Most of their money is for providing Enterprise services. The company I work for (50k+ employees) uses it, and I've talked to quite a few people in other Enterprise sized organisations that also use it.
Last lot of figures I've seen shows $140m in profit in 2016, up from $90m in 2015, and $70m in 2014. Assuming that growth continued, they are likely making $200-$250m profit per year currently.
I rashly suggested most people weren't paying for GitHub (sorry) and your numbers astonish me. I don't doubt them, I'm just amazed. Why would anyone pay non-trivial sums for something they could set up and run themselves? Is it sheer laziness?
I'm beginning to think that the Age of Stupid will reach its asymptote of cretinism when the Internet of Shyte literally offers an app-gadget to wipe people's arses.
There are lots of things I could do myself - grow food, supply my own water, mow my own lawns, catch my own mouses - but choose not to because there are lots of other more fun things I would rather do instead. My cat would be annoyed if I caught her mouses. Didn't know I was being stupid. Sorry!
"... for something they could set up ..."
Who's "they"? I can certainly set up something on my home computer here, but that's not going to help my sometimes collaborators in Canada and France (who, by the way, only discovered my project by searching/browsing GitHub repositories in the first place).
It's not just the version control. Git can be set up on anything. It's GitHub's collaboration features that made it such a powerful tool.
I'm the AC in the 50k+ employees who posted the numbers earlier.
When we were setting up the GitHub, I had a chat with one of the team who were managing the roll out. Asking things like why GitHub etc.
Bearing in mind we have several DCs around the world (2 or 3 depending on region), so we have places to host our own services and the expertise to do this ourselves. Git itself itself isn't difficult, but obviously it needs hosting, it needs resilience, security (this is for internal commercial projects, not Open Source). etc etc.
Bottom line was cost. Once you took into account that someone would need to build and maintain this, and it was being rolled out to all developers in our org, so needed to scale etc. It turned out far cheaper per head to get Enterprise GitHub, with a floating license, than it would have been to do it ourselves,
Plus most of our developers were already familiar with GitHub, due to people contributing, or just looking at, Open Source projects, so there was no real learning curve, just an email saying you now have Enterprise access to GitHub.
git != github.
github gave us the pull-request, git cli will print something to stdout but github makes the whole process pointy clicky. They made the fork / edit / pull request thing possible from a person without commit rights to your project. This is a key feature, gitlab have it too now, neither are trivial to setup and manage. Github then wraps a social media thing around contributions, its far more than a repository.
If you install github locally or an opensource alternative you don't get "enough eyeballs" on your code. Microsoft is buying the ruby code and the eyeballs.
"Last lot of figures I've seen shows $140m in profit in 2016, up from $90m in 2015, and $70m in 2014."
Those numbers were *revenue*. Not profit.
"they do charge money for private repositories"
and it seems to be cost-effective for small development projects in which you have one or more people working from home (or even more remotely). it's actually pretty good that way, as you can use issues and pull requests to manage collaboration, and the wiki to manage documentation.
I have a previous customer that managed their project this way. For the most part, it worked well. When done right, you could test a change in a branch, then commit it, and do 'git pull' and re-build things [as needed] in a very short period time, to update the production system. Down, git pull, build, up. pretty fast. And private repos, so the 'secret sauce' was kept secret [still had open source but it was in the form of taballs that were snapshots of firmware releases, not day-to-day history visible to the public].
It isn't worth that much. But the ability to further manipulate you might be.
I think the real question is—
What is Microsoft's strategy to extract that much value from GitHub?
Pridefully foolish as Microsoft's strategic decisions so often are—El Reg readers can probably list half a dozen blatantly stupid screwups off the top of their heads, and not all of those are from the days of arrogantly halfwitted Ballmerups—the intention is to get surplus value from the money paid. Thus the horrible LinkedIn (it isn't 'necessary' at all, BTW: it's much nicer to get work by word of mouth) wasn't really about improving the value of the platform, it was about buying eyeballs and thereby entrapping yet more people into the MS 'ecosystem'*¹. Thus the Skype experience: MS wanted the users, had no interest in improving the product for anyone, cares only about the captive userbase ... which is why it keeps getting worse.
So you must ask, how will MS try to extract that enormous value—not from expanding or improving GitHub—but from you, the users? You do not pay a fee for GitHub, do you? And you've no intention of ever paying. So that $7.5bn is coming out of your skin. MS is willing to bet that it knows how to squeeze the cash out of you.
That's the modern internet-giant business approach. Customers are now the product to be sold, rented, hired, wrung out and have their privacy prostituted at every turn, so the top priorities are—
• How do we catch them
• How can we rape their wallets
• How can we sell them
—and, super important—
• How do we trap them here?
*¹ Ecosystem = Punji-stake-pit
I don't know how much money it would have been worth before MS bought it, but with the Anti-Midas in charge, it will likely be less in short order.
I wonder, we talk big about how much we dislike MS - I do as well, though not as much as some. How will this work out? General rule of thumb is that in case like this, most of the public does not share the outrage of the engaged community and the unwashed masses continue to graze contentedly.
Win 10 is not a great success, true, but then again most stores still sell PCs with it, companies run it, etc... Apple is pricey enough that Windows stays in the running and are not to everyone's taste either. Linux is still patiently waiting on its desktop year.
The sky didn't fall for Win 10 + telemetry, not even for 8. Face it, some of us care, most people don't.
But Github is different: its constituents are precisely developers, open-sources ones to boot. Technically, individuals could easily move, if they knew where to go. Companies maybe less easily, but one thing MS has achieved is giving Google a black eye for relying on Github - at least some big organizations will be motivated to fly the coop. Unlike Linkedin (or FB), Github has limited built-in network effects - the apt-get/pypy/npm/yarn/etc... type endpoint installers can be just rerouted to look elsewhere - they're their respective ecosystems network-effect bits, not the hosting providers. Many devs may take considerably less pride in having their calling cards on MS properties.
We're on Github because we choose to be, not because we have to.
If this ends up turning into a referendum on the popularity of Microsoft with developers in the late 10s, what will it tell us? What will it look like if, in 2-3 years, Github is now #2 or #3 in growth?
I am very doubtful about the upsides for MS, but a severe drop in Github influence and market share from now on would be a gigantic and very public egg in the face for MS, bought at the cost of $7.5B.
"Does it sling ads at people?"
Probably it will - not hard to just use an ad blocker.
"Charge money for businesses to keep their code there?"
Of course if they want private repositories and any level of support. Just like now.
I'm sure they will (subject to GDPR optin of course) try to sell you stuff such as developer tools and related services too.
How can it possibly be worth that much?
It does not. And neither do most Californication endeavors. However, the way taxation in USA is set-up it is significantly more efficient for a large company to spaff a few Bn up the wall every year in M&A than to spend the SAME amount on internal R&D. In fact, no company in software, telecoms, networking as well as most other high tech industries can spend a few Bn to develop a product. The beancounters will not allow it because the stock market ambulance-chasers will disembowel it the moment they hear about it.
As a result this vastly overvalues the valley scene and feeds a never ending circle of "innovation" in it.
> How can it possibly be worth that much?
1) Developers: No one who hasn't lived behind a rock the last 15 years would ever again develop against a Microsoft API. Their platform and their brands got burned. So Microsoft is seeking to brainwash young devs.
3) Private repositories: Many companies outsourced the Git repo hosting to Github. They pay to have it hosted as a "private repo". Guess what. Now Microsoft has full read access to all their competitors source code (minus the big 4 Google, Facebook, Apple, Amazon; but all the rest). So today Microsoft is lacking behind in several gartner Quadrant, in 2019 Microsoft will have magically be the "leader" by simply knowing all secrets by looking through all competitors source code! Given that Microsoft was in talk with Github for more than two years, rest assure that all source code that was on Github starting in 2016 has been kept on backups, even if it was marked as deleted. They don't care about GDPR, etc at all anyway. They are too big and have too much money. Unfortunately, Microsoft hasn't been split up in the 1990s, they are still a huge monopoly and a very bad actor.
Shouldn't they sit in jail incl Bill, for evil anti-competitive behaviour over so many decades?
Their lawyers will make damn sure they don't look at others' private source code, because it would be VERY easy for a company to sue them if that code was misused by Microsoft because proving they had access to it would no longer be necessary to the case!
If I were remotely interested in making jokes that rely on "well known Internet meme[s]" I wouldn't be doing it here. I come here, and to the comment threads in particular, because it's full of people who know all sorts of interesting (and sometimes obscure) stuff that I don't. And sometimes they make jokes about that stuff. And if I'm interested I look it up* and maybe have a laugh. Or, you know, I don't and move on to the next comment.
*like say typing "github lesbian" into a well-known search engine.
Biting the hand that feeds IT © 1998–2019