
Canonical’s Ubuntu may have that special je ne sais quoi when it comes to the desktop, but will enterprise users be similarly enchanted by its new commercial support?
That’s the question of the moment, as Linux bloggers ponder the company’s recent foray into Red Hat territory with its new Advantage enterprise service.
“Ubuntu Advantage provides systems management, support, legal assurance and direct access to the experts,” in the company’s own words. “Available for both servers and desktops, you can choose the right service level to meet your needs.”
‘Just Pay Us Some Money’
Now, “when it comes to the commercial sector, there’s no question that Red Hat is the king,” as Linux Magazine’s Christopher Smart recently pointed out. “Canonical might not be deliberately setting out to challenge Red Hat with their new support offerings, but they are now competing in the same space.
“So with this latest move from Canonical, could the crown be possibly up for grabs?” Smart wondered.
Over at LWN, meanwhile, jake quickly picked up on one surprising feature in Canonical’s offerings: the availability of indemnification in case of patent claims.
“We take care of intellectual property (IP) infringement legal claims brought against customers in their use of Ubuntu,” the site’s Assurance section explains.
“Hmm, that’s curious, isn’t it?” jake wrote. “In one way, it’s sort of playing into the whole racket. ‘Ubuntu Linux could be at risk, so just pay us some money and we’ll make sure that you’re covered.'”
‘They Will *Surely* Tarnish Linux’
Linux bloggers, as is their wont, had no shortage of thoughts to share.
“Ubuntu’s indemnity is the same offering as Redhat and other commercial vendors,” wrote sharms in the LWN comments, for example. “They all offer it because it is required for doing business with actual clients. Nothing to see here.”
On the other hand: “That is wrong,” countered rilder. “It is like indirectly stating that if you don’t (or use competitor’s linux) you can get ‘sued’ for using linux. They will *surely* tarnish linux with that.”
It quickly became clear that the topic was a divisive one. Linux Girl took to the steamy streets of the blogosphere for more insight.
‘This Scatter-Gun Approach Won’t Work’
“This makes me think of jam. Raspberry jam,” said Barbara Hudson, a blogger on Slashdot who goes by “Tom” on the site.
“In The Secrets of Consulting, Gerald Weinberg talks about the Law of Raspberry Jam: ‘The wider you spread it, the thinner it gets,'” Hudson explained.
“Canonical simply isn’t credible when they are running in so many directions at once — tablet, virtual, netbook, cloud, desktop, server,” she opined. “This scatter-gun approach won’t work.”
‘The Death Smell’
“I call it the ‘death smell,’ because beneath the glossy surface, you can almost smell the sweat of the stand-up comic who is bombing because the audience isn’t going along for the ride, or the salesperson who starts babbling when they sense they’re losing the sale,” Hudson asserted.
As for Canonical’s “patent protection” promise, whereby it vows to “replace or modify” any infringing software, “some quick statistics show that Canonical simply doesn’t have the expertise to do much rewriting,” Hudson noted. “Canonical has never been a significant contributor to linux.”
In short, “if Canonical were the only source of support, I’d move to one of the *BSDs for my own safety,” she concluded. “As a bonus, freeBSD’s retro text-mode ascii-art demon boot screen isn’t nearly as fugly as Ubuntu.”
‘It Is Time to Kick M$ Out of Servers’
Legal threats “should become less of an issue after SCOTUS rules on Bilski soon,” blogger Robert Pogson predicted. “With software patents muzzled, legal insurance should be a non-issue.”
That aside, Canonical’s move “is not as much a competition as an expansion into a growing market,” Pogson opined. “GNU/Linux has taken about all it can from Unix operating systems; now it is time to kick M$ out of servers.”
The “ridiculous complexity and overhead of an OS designed for a single-user desktop has no place on a server, and both Red Hat and Canonical can ramp up share for GNU/Linux desktops/thin clients/virtual desktops and services,” Pogson explained. “With that other OS reduced in share on the desktop, there is far less need for that other OS on servers to manage the mess M$ created.”
Customers will get “better service for less money,” in fact, “using GNU/Linux on servers with help from either Red Hat or Canonical,” Pogson concluded.
‘They Pay Good $$$$ for Support’
Indeed, “having commercially supported distros with proper package managers would lower my stress level considerably,” Montreal consultant and Slashdot blogger Gerhard Mack agreed. “Dealing with RPM-based distros has a tendency to make me want to throw things through my monitor.”
Finally, “this is simply a sign of something I’ve felt will happen all along: Canonical will eventually get out of the desktop game,” Slashdot blogger hairyfeet told Linux Girl. “Trying to compete with MSFT in the desktop, especially when you don’t have the $$$$ to hire your own development team to ensure QA, is simply madness.”
On the other hand, “businesses can save money using Linux on servers,” he pointed out, “and they pay good $$$$ for support.”
GNU/Linux is not just for servers. The same advantages of modularity, openness, standards-compliance, and Free Software licences apply equally to the desktop or the server. With cloud computing, thin client/VDI, and even good old X, the desktop is just another service. Software developers, IT managers, end-users, and the bean-counters all benefit.
Canonical has given a lot to GNU/Linux. Debian GNU/Linux and Ubuntu often exchange ideas, code and bug reports. Ubuntu has some good ideas about user interfaces and marketing. They are a huge asset to the Free Software community and they will be a growing asset to businesses that use their services. We need all sorts of individuals and organizations involved in GNU/Linux. It makes us all stronger.
Sorry Pogson, but I have to disagree about the desktop. Why? Because the server hardware is for the most part standardized around well supported by FLOSS chipsets, whereas desktops are this huge mishmash of proprietary chips that many will NEVER have any documentation for, which means Linux WILL fail there.
It is simply a matter of numbers: There simply aren’t enough developers to insure that new devices don’t break old devices, that devices that worked in Distro Foo work in Foo+1, and with everything from the kernel on up being as the shifting sand with regards to stability EVERY device needs QA, that simply won’t get done. Look at Ubuntu forums and see how many "Update Foo broke my hardware" you see. Despite the FUD that is often spread, that simply doesn’t happen in Windows and OSX.
With servers you have a MUCH smaller subset of vendors, many if not all are paying serious $$$$ to ensure their devices have good Linux support, because in the server space it is a good ROI, it simply isn’t on the desktop. If you ever get a chance, check out the Dell Ubuntu netbooks and notice the repo setup. Notice how they disable the default repo and put in their own which almost never updates? That is because desktop updates BREAK hardware in Linux, and that just doesn’t happen in servers nearly as much thanks to $$$$ being spent by OEMs.
Sorry to burst your bubble friend, but like everything else in this world it comes down to $$$. There simply isn’t enough ROI to bother with Linux support on the desktop, so the vast majority of hardware OEMs don’t. Just the opposite is the case in servers. that is why for the most part Linux "just works" even on brand new servers, whereas on a brand new desktop you’ll be lucky to get 60% hardware supported.