laugh or cry
not sure
SQL Server running on Linux, with embedded R and Python, is Microsoft's most successful SQL Server product ever, said JG "John" Chirapurath, general manager of Azure Data, in an interview with The Register at Build 2018. "It is the most successful server product we've ever released, in terms of downloads," he said. "Our …
Presumably because its free and its not Oracle MySQL.
Microsoft know the vast majority of users will eventually want features like always on availability group clustering in production so they will eventually need SQL on Windows Server and then end up paying for the fully capable version.
I would say laugh...
As Ghandi said:
First they ignore you, then they laugh at you, then they fight you, then you win.
I think Microsoft have gone through all the stages with Linux.
Can't remember the laughter part, but Ballmers 'cancer' jibes were certainly fighting talk.
"Microsoft know the vast majority of users will eventually want features like always on availability group clustering in production so they will eventually need SQL on Windows Server and then end up paying for the fully capable version."
Thats entirely the point. Get them hooked then reel them into the windows ecosystem when their systems are too complex to switch to another DB. In supermarket terms it would be called a loss leader.
Microsoft know the vast majority of users will eventually want features like always on availability group clustering in production so they will eventually need SQL on Windows Server and then end up paying for the fully capable version.
1. You can do Always On Availability Groups on Linux.
2. If you're using it for production data, you still need to license SQL Server on Linux, same way, same price as SQL Server on Windows.
3. Microsoft has stated that their long-term goal is feature parity between SQL Server on Linux and SQL Server on Windows. Many of the features not supported on Linux are features that have been deprecated on the Windows version for some time, so I expect those features to fade away in Windows rather than be implemented on Linux.
Word and Excel can both be used on Linux through portal.office.com. It's unlikely there will be a native port of the code at this point, given the low uptake of desktop Linux. Server Linux is very popular though, hence porting SQL Server. It did help that SQL used almost no Windows API calls too :)
"No that's not the case at all."
Maybe you think it uses magic strings? On Windows server, it uses Windows APIs, and on Linux it uses Windows APIs thunked to the Linux kernel APIs. It's not like say Oracle which has Oracle only proprietary raw file systems and APIs.
Quote:-
Webapps != full MS Office. It's like Google Apps - a half arsed cut down product.
Which ironically is probably what at least 90% of Orifice users need from a functionality POV. The rest is just "meh" and "so what".
Hey MS, how about releasing a version of Office with the functionality and appearance of Office 2003 that includes all the patches but not all that gancy gizmo's in Excell or the Ribbon?
I'm sure that it would sell very well even at half the price of Office 365.
I suspect an accurate usage survey would show about 45% of the features Office 2010 being used by about 95% of the users. Another 10% of the features used by 4.5%+ of the users and the rest of the features each used by a small scattering of users with no definitive trend.
The people who use font changes to indicate section headings and spaces to centre them, manually type out a Table of Contents, add up columns in a spreadsheet with a calculator and leave the document formatted for US Letter paper, are hardly going to notice.
entirely down to the laissez-faire attitude (possibly with a soupcon of condescension) of the Linux community steadfastly refusing to concentrate on Enterprise level desktops.
The lack of an industrial strength Linux equivalent for Outlook is embarrassing, 12 years after I told my colleagues Linux was ready for the desktop.
".... when they create a Windows Subsystem for Linux that lets developers run Linux environments -- including most command-line tools, utilities, and applications -- directly on Windows, unmodified, without the overhead of a virtual machine."
And without the overhead of the Linux kernel. Wait, they already did that!
I have copies of their news postings on Usenet from that time. That’s probably why they were wise to secure the Xenix franchise so early.
Still doesn’t explain how Gates missed diving into the Internet in his book The Road Ahead. (At least, in its original editions.)
Hugo, they owned Xenix, which was a early Lunix port. From memory they never implemented record locking, so it didn't attract that many customers. Nortel in Hemel Hempstead (Northern Telecom in those days) built a 16bit (can't remember the name of Intel's 16bit bus architecture that competed with VersaBus and VME), Xenix system in the early 80s. 16MHz 8086 CPU I seam to recall.
I work in a mixed OS environment with BSD, several Linux flavours, MacOS and Windows.
In my experience, the Unixes (and Unix like OSs) differ slightly less than human beings do from person to person.
By comparison Windows is further away genetically than the great apes, even though they have a common origin.
"doesn’t explain how Gates missed diving into the Internet"
Gates was never a seriously visionary techy, but he knew how to broker a corporate deal or two, and to build a Microsoft-dependent ecosystem (at every level from Dell/HPQ/etc's ongoing willingness to pay the Windows tax on desktops and laptops, to a global community of backroom Certified Microsoft Dependent Associates). Does that help explain?
"Gates was never a seriously visionary techy, but he knew how to broker a corporate deal or two, and to build a Microsoft-dependent ecosystem (at every level from Dell/HPQ/etc's ongoing willingness to pay the Windows tax on desktops and laptops, to a global community of backroom Certified Microsoft Dependent Associates). Does that help explain?"
I'll leave the argument about Gates being a visionary techy to one side. Gates certainly understood the businesses that used MS products and tried to steer the products in a direction that met their needs at least as well as the competition. Given his some of his successors products, he was certainly more visionary...
The rest of the Microsoft model was very much in-line with the rest of the industry in terms of doing things that others already did, but cheaper.
For MS certifications, throwing cheap/free software at techies to give them the chance to run the products MS was trying to sell without too many concerns about licensing arguable led to the boom in IT. While it benefited MS, I think the legacy is much broader than just dodgy paper certs.
> and tried to steer the products in a direction that met their needs at least as well as the competition.
You are confused. Microsoft succeeded by _buying_ the competition and claiming that they were Microsoft products. They also bought competition in order to kill them, or did so through incompetence. MS-DOS, Xenix, IE, FrontPage, Visual BASIC, MSC, Powerpoint, and many others were bought.
The only reason that Microsoft hasn't been able to kill Linux by buying it is because it can just be forked and continue.
"MS-DOS, Xenix, IE, FrontPage, Visual BASIC, MSC, Powerpoint, and many others were bought."
Add AutoRoute to that list, along with many others. AutoRoute went rapidly downhill after it became an MS product rather than a Nextbase product.
Plus there's more than one way for MS to 'buy' a product. Add Windows NT to the 'developed outside MS, profits made by MS' list, and note that people like Dave Cutler (architect of what became Windows NT) is still an MS employee.
Also still funded by MS are people like Professor Tony Hoare of CSP fame:
https://en.wikipedia.org/wiki/Tony_Hoare
> Add Windows NT to the 'developed outside MS, profits made by MS' list
Actually DEC threatened to sue over Cutler NT and MS settled for an alleged 100million plus keeping Alpha versions and joint promotion of NT and VMS.
http://www.itprotoday.com/management-mobility/windows-nt-and-vms-rest-story
@Richard Plinston
Thanks for that, I'd seen most of the content some time ago, in various incarnations, but it bears repeating, especially given who the author was (in 1998) and where he is now (in 2018).
Who's the article author? A chap called Russinovich, once widely known for his independent Wininternals and Sysinternals work. In 2006, MS biught the company. Where is Russinovich now? A very senior MS employee (CTO at Azure?).
Gates knows how to do deals.
See also:
http://www.nytimes.com/1998/09/10/business/digital-employees-tell-of-threats-by-gates-over-product.html
"Mr. Gates had told Mr. Palmer, 'You have to decide if you're Larry's friend or my friend,' '' Mr. Chaiken recalled Mr. Supnik's telling him. ''They looked at the revenue that Digital received from its relationship with Oracle and the revenue it received from its relationship with Microsoft and figured out which number was bigger.''
We know who Gates was/is. Palmer was top man at DEC/DIGITAL/etc, even more forgettable than Oracle's Larry Ellison.
What the powers that be in DEC HQ had apparently forgotten was that a board level promise from MS was worth its weight in gold (ie nothing).
Gates knows how to do deals.
As Todd might once have written, get your shit together be a real man :)
Multibus/IEE796?
From the era of Intel's iSBC hardware and iRMX software, and blue boxes, maybe?
Not part of the Wintel plan for world domination, hence vanished without trace from most people's recollection.
I don't remember a made-in-Hemel, by Nortel, computer matching your description, but in the years before their collapse, Nortel did end up owning a surprising number of companies in the UK, with some thirty thousand UK employees affected by the collapse/theft of the Nortel UK pension fund ten years ago (and still no clear final verdict, and not much coverage here or in the mainstream press),
Yep, there was a system manufactured in HH by Northern Telecom - (The Project) Vienna system. It ran on 80286 (I was part of the European support group for the product based in HH during the mid-1980s). From memory it supported a maximum of 32 serially connected users on very little RAM. Originally it ran Xenix III (based on Unix System III). The last version was SysVR2.
The nice thing about working there was that we had access to the complete Xenix source code.
As for the pension fund, I was thinking about the other day .....
Microsoft has never been shy of adopting new platforms, once they reach some minimum threshold of seriousness. "Embrace and extend" flows inevitably into "extinguish" iff Microsoft's additions make the existing platform significantly more attractive to significantly more people.
If people start switching to Linux purely because they can now use SQL Server on it, it'll be in trouble.
"Microsoft has never been shy of adopting new platforms, once they reach some minimum threshold of seriousness. "Embrace and extend" flows inevitably into "extinguish" iff Microsoft's additions make the existing platform significantly more attractive to significantly more people"
For crying out loud, give that ever so outdated "Embrace, extend, extinguish" rubbish a rest. Linux has effectively won on everything but the desktop including supercomputers, internet of things and so on so it is impossible for Microsoft to challenge or eradicate Linux.
The last time I checked, over 40 percent of all virtual machines in Azure were running Linux. That is a effectively a sign of Microsoft's weakness because they now need Linux in order to make money. Ballmer has been well and truly defeated and Microsoft under Satya Nadella is learning to live with Linux. The war now only continues in some people's heads and not in real life.
"The last time I checked, over 40 percent of all virtual machines in Azure were running Linux."
Well overlooking the fact that ALL of Azure itself runs on Windows / Hyper-V, that's largely because hardly anyone trusts production environments to the public cloud so far. These are mostly development and / or web based environments.
However that being said, I doubt Microsoft care about making money from Linux. If you think that they don't have to support the OS directly and there is little Microsoft R&D required, then it's effectively more profitable than Windows VMs!
Seriously? SQL Server will be in trouble if it ever relies 100% on Linux, but trust me there is no way Linux can be stamped on by MS. I'll eat the brickwork in my house on the day MS kill Linux!
If MS try to stamp a particular distro and get even close to killing it, it will simply get forked and pop up elsewhere. Linux and FOSS in general, is like a balloon full of water, if you squeeze some part of it, it simply pops out somewhere else having been forked.
That's what occurred to me. How many are docker pulls and other container tech, that gets pulled and never even started.
You can pull SQL Server for Linux for free but you still need to agree the license and where required ensure you have a paid up membership to the MS clubhouse to use the Std and Ent versions.
I could download, build and install MariaDB, for nothing; and get the complete Source Code, which I can subject to an independent security audit. If it doesn't quite suit my needs, I can employ any competent C programmer to fix it.
..... Or I could spend a lot of money on a pre-compiled Microsoft SQL Server, not get the Source Code, have to trust them without a security audit, and if it doesn't quite suit my needs, I'd better change my needs.
Hmm, tough decision!
"I can employ any competent C programmer to fix it."
Is a very expensive option, that's why most companies go with a commercial option.
I really depends what you want your DB for, if all you're doing is using it as a back-end data store for a web application with < 10,000 simultaneous users you can use almost anything. MSSQL is good for some things, MySql/Maria/Postgres for others and Oracle is terrible regardless (troll face). Heck, some things you're better off with an object store like MongoDB or Couchbase.
My point here is that it's very dependent on what you're doing, what the best tool for the job is. Sometimes commercial software makes a lot of sense and sometimes it doesn't.
And when your C programmer decides to walk at short notice and you discover that it'll take 6 months for someone to reverse engineer the changes and the tools used are no longer supported..... Not an uncommon scenario. If you've got the resources to deal with that then that's dandy, but if not...
Horses for courses
One place that may be the origin of a lot of downloads is that you can use the MSSQL server docker image to do development against a copy of MSSQL on your workstation in order to test code that uses the DB cheaply and easily before pushing code to staging servers on Azure. (Just set it to "developer edition", which it defaults to, and DON'T DEPLOY TO PRODUCTION because the EULA expressly forbids doing that.)
At least, that's what I'm using it for. It's really nice because you can a) run it on a Mac via Docker-for-Mac, b) use Docker's functionality for snapshotting the entire SQL server state, for repeatedly testing destructive operations, since SQL Server doesn't currently support that very well AFAICT. The slowest part is waiting about 6 seconds for the SQL server daemon to load and become usable.
since when? Win-10-nic says the opposite.
At least their SQL server group is STARTING to "get it". But for the rest of the company, it's same thing, different day. Customers are "minions", to be herded and coerced into doing things "Microsoft's way" so that the market can be cornered, dependency can be established, and nobody DARES go outside of the box!
And YOUR computer is under THEIR control!
"At least their SQL server group is STARTING to "get it". "
Apparently the Windows 10 group get it too. You can now run Linux apps directly under Windows without the hassle of a VM, a separate file system or the need for a Linux kernel.
Windows Server 2019 will likely run Linux Docker containers without a Linux kernel too. One OS to run nearly everything will be an attractive proposition to many.