Posts by thames

Pinning a value on big tech's top names. Not as easy as it looks

Random Monkeys

"Repeated experiments have shown those random monkeys outperform most fund mangers."

Repeated experience have also shown that random monkeys can outperform most "tech" pundits when it comes the future of the industry.

You'll notice that some of the companies in your list are highly diversified and operate in mature markets, while others are narrowly focused and operate in rapidly growing markets.

If you primarily sell PC operation systems and office packages, then your room for growing sales of those products is limited. If you are primarily in e-commerce and "cloud", then rapid growth comes naturally. There needn't be anything extraordinary about the companies in either case, they are just following the general trend for the market they are in.

As for how does a company get in on the ground floor of a growth market? Very often, it's by chance. All they need to do then is to not screw it up.

• IBM, Microsoft, Oracle - Diversified companies with slow growing products. It will be hard for any of them to come up with a major new product line to replace their existing ones and give them rapid growth. Microsoft's failures in the mobile market are particularly notable. All are trying to get on the "cloud" bandwagon, but that wagon is a looking to be a prettry crowded one.
• SAP - I can't explain that one.
• VMWare - They've been focused on one of the hottest markets in IT - virtualization. There is also some financial engineering going on with respect to their share ownership.
• Google - They're big in the rapidly growing web services and mobile markets. They have a history of producing a lot of failures, but they also have a history of ruthlessly cutting failed projects and piling effort into successful ones. In other words, they throw darts at a dart board, but the markets they focus on are ones which let those darts pay off very well and very quickly when they hit something.
• Amazon - E-commerce is a hot market. They're dominant in cloud infrastructure, but I'm not convinced that's a big factor.
• Salesforce - Somebody, one of the big but stodgy vendors, will buy them out at a good premium.

The BIG stretch: Software and flexing your firm's size

"enterprise software is there to ..."

I would say it's there to do for the office what machinery does for the factory floor. It's a form of automation. The problem is that a lot of it is just very badly done. Failed machines get fixed or scrapped. Failed ERP systems live on forever bleeding away time, money, and the will to live.

I think the big problem with enterprise software is the lack of tangible and measurable benefits in most cases. You know if a new factory machine has saved labour on a production line. You just have to count the number or workers and compare that to output. ERP and CRM systems often seem to get installed with no clear goals or way of measuring the result.

As for the size of firms, the reason why the largest companies tend to be the big customers for "enterprise" software is that the benefit of this software is often so small that you need large economies of bureaucratic scale to see any net benefit from the up-front cost of implementation and the ongoing license fees. When someone figures out how to do this sort of thing quickly and cheaply, we'll see it filter down more to smaller businesses.

Microsoft points PowerShell at Penguinistas

Re: Ha! Ha! Ha!

@h4rm0ny - "It's depressing that nine people so far have modded you up for that."

It's depressing that 18 people have so far modded you up without looking at your link. That's just a link to a page on how to download and compile the source code. "Yum" is used in that process to install the compiler, Python, and Open-ssl libraries, assuming for some reason they are not already installed. Yum of course is the Red Hat RPM installer, used to install software from the Red Hat / Fedora / whatever repositories.

After that you have to download some more source code from a third party web site (opengroup.org) and compile that. Amusingly Microsoft points to a dead link. I guess you are expected to rummage around on opengroup's web site to look for it.

Assuming that you have compiled everything from source, then you need to cobble together a sys V init script. They helpfully give a sample which may or may not work. No Systemd or Upstart files or tips though.

All in all, it's a prime example of the sort of thing which Windows fans claim is the normal install procedure for software on Linux, and which actual Linux users wonder WTF these Windows fan boys are talking about. Now we know. It's apparently the sort of thing which they have come to expect from Microsoft.

Let us know when there's actual Linux packages available for actual Linux distros, and perhaps we can take a look at it.

Ha! Ha! Ha!

I just downloaded the package from Microsoft to have a look at it. The documentation is a "docx" file and the package for installing on Linux is an "msi" installer. Me thinks Microsoft haven't quite got the hang of this "Linux" thing yet.

"Whether those same admins are willing to move to PowerShell, instead of the the familiar tools they us to manage Linux, is another matter."

I doubt that LInux/Unix server admins will have the slightest interest in it. Existing management tools for those platforms are well developed and established. You can use things like Puppet to manage Windows servers as well as Linux servers, and it's tools like that which Linux/Unix admins will turn to first.

What I suspect is this intended for is an attempt to try to keep Windows based management tools relevant in the data centres of existing customers. If you can use Linux based tools to manage Windows servers but not use Microsoft's tools to manage Linux servers, then Microsoft would get squeezed out of the market.

I expect these tools will be accompanied by some additional Windows only software which they can sell as an add-on. It's a pretty common business model.

I'm waiting for our anonymous marketroid to chime in with his spin cycle about how the fact that Microsoft has to strive to work with Linux systems still means that Microsoft is destined (someday) for total dominance in the data centre.

All-Russian 'Elbrus' PCs and servers go on sale

Re: Will the real instruction set please stand up?

According to Wikipedia, this is either the same organization as made the mainframes or a descendent of it. They made several lines of computers. The originals had some similarities to Borroughs mainframes, but they were independent designs, not copies. In the 1980s, they switched to a VLIW architecture. This current chip seems to be a descendent of that.

They also made a separate line of products using the SPARC architecture. Something that isn't clear is whether this new chip somehow incorporates the SPARC instruction set as well as the Elbrus VLIW one, or whether that is just garbled reporting mixing up the two product lines. I suspect it's the latter.

Their chips are (according to Wikipedia) used in the "space program, nuclear weapons research, and defence systems". Now you know why the Russians use them. The Americans won't sell chips to them for use on those markets.

The model of chip that comes out this year is the 8S, which is 8 cores, 1.3 GHz, and built on 28 nm. It sounds pretty respectable even by today's standards.

That's not the point

"If MCST's promise of compatibility with other operating systems and code written for them is accurate, "

It can run Linux. That's all they need. They're not going to run Windows commercially since running Windows and associated software would just defeat the whole purpose of the project, which is an independent technology base.

"its conceivable Russian outfits worried about NSA backdoors in Intel and AMD kit have themselves a platform on which to get some decent work done."

The US restricts what models of Intel and AMD CPUs can be exported to various countries for various purposes. With the Elbrus, the Russians can ignore the American "chip blockade" and do whatever they want. This is the same reason the Chinese are building their own CPUs based on MIPs designs.

"Running complex workloads looks rather less likely."

Eh? It should run pretty much any open source software plus whatever Russian proprietary software they decide to port to it. The exceptions which may be problematic are ones like Java or C#, since those would require writing code generators for the JIT compilers, which may not be worth the effort. Anything written using anything else, C, C++, Fortran, Python, Ruby, PHP, etc. should not be a big deal.

All the American proprietary vendors would be given the boot from government installations, so the fact that their software isn't "supported" on an Elbrus chip is irrelevant. If "complex workload" is code for "American proprietary vendor who buys me pints at trade shows", well getting rid of them is the whole point.

Really, the only issue here is to what extent the Russian government is actually interested in Elbrus chips and to what extent it's just PR from MCST talking up the sales potential. There was another Russian company who not too long ago was saying they were going after the same market using 64 bit ARM server chips. Only one of these at most is going to get the nod.

The video linked to in the story seems to be from some blogger, not from either MCST or the Russian government. He's used some video from Elbrus and added his own commentary. God knows to what degree it reflects reality.

NetSuite's leap over to Azure cloud - a shot to the pills for AWS?

Is this the origin of the Salesforce rumours?

Is this perhaps the origin of the rumours that Microsoft was going to buy out Salesforce? Perhaps what Netsuite is really doing is positioning themselves for an acquisition by Microsoft. They may have an agreement that if they move to MS Azure and hit certain targets, then Microsoft will buy them at an agreed price.

It's certainly a viable end game to 17 years of no profits. Microsoft would buy a business like this, but Amazon won't. Microsoft on the other hand wouldn't touch them if they were still running on AWS.

Microsoft is already in the ERP/CRM business with MS Dynamics, which was formed from a couple of acquisitions. This buys them some more market share, and it's a company which is already cloud based, which fits in with Microsoft's business direction.

As for "why did they jump", well they haven't jumped yet. They just have a plan to jump. it will be interesting to see how well that plan goes in reality. A lot of people are nervous about "cloud" because they are afraid of vendor lock-in. If this move goes ahead and works well, some of that concern may be alleviated. On the other hand, it will also put other large customers in a better bargaining position than they are now since they would have a more credible threat to switch clouds.

I will bet however that what Netsuite will be paying Microsoft bears no resemblance to what you or I would pay. Microsoft may be paying them to host there, for all we know. It's like how Microsoft paid Nokia to use Windows Phone ("platform support payments"). Nokia sold the handset business to them after the payments ran out.

It's a picture of what NetSuite is going to look like a couple of years after they've finished migrating to MS Azure and Microsoft sits down at the table with them to negotiate the new T&Cs from a much stronger bargaining position.

HORDES OF CLING-ONS menace UK.gov IT estate as special WinXP support ends

Re: The NHS

@Boris the Cockroach - "problem is much like mine in the industrial arena."

I was involved in commissioning a new machine in a factory a while ago. It used a PC running Windows 2000, which at the time had just had support terminated. I asked the rep from the company which made it about that, and he was shocked and dismayed to hear this. It was their latest model. They had just got the design to work properly. Damn this Microsoft for doing things like this to them. Anyway, it wasn't their responsibility, since they just buy the PC from someone else and stick it in their machine ($100,000 plus) and load their software on it. Shrug, what can you do anyway? Now let me show you the interface to the conveyor system. There were only two or three companies in the entire world which made that sort of equipment, and the others weren't any better.

Oh, and the biggest company in the industrial automation market (who shall remain nameless, but they're a giant whom you've all heard of) had one of their premier software products running on CP/M in an emulator which ran on MS-DOS, which ran in a compatibility box on Windows. Finding a PC which still had a serial port which was compatible with it (USB serial converters didn't work) was a pain in the arse. A very large chunk of the world's factories, utilities, water systems, etc. was run by their hardware.

Long term support for our software? Oh, yeah, we've got this new product you ought to buy. Just rip the guts out of your factory (power plant, water supply, etc.) and replace it all with our new stuff.

Microsoft's secret weapon in browser wars: Mozilla's supercharged Asm.js

It's probably better to think of Asm.js as a form of intermediate representation. A lot of compilers go from source code, to intermediate code, to machine code. Among other advantages, this lets you combine various different front end language translaters with different back end machine code generators. In the case of Asm.js, the intermediate representation just happens to be a sub-set of Javascript.

The Pypy Python JIT takes a somewhat analogous approach in that it uses a subset of Python called RPython which can be highly statically optimized into machine code by the compiler. I say "somewhat analogous" because they don't use it as a compiler target for writing user supplied programs. Instead, it's just part of the toolkit they use to write interpreters and JIT compilers in, including Pypi, along with PHP, Ruby, and others (it's a general purpose language toolkit).

The basic idea behind both is that speed has nothing to do with "compiler" versus "interpreter". Both Javascript and Python are compiled. The real issue is "static" versus "dynamic". Both Javascript and Python (along with Ruby, PHP, and various others) are very dynamic languages which can only really be analysed completely by observing them at run time to see what they are actually doing. However, if you strip out the dynamic features from them and leave a core which can be analysed statically, then you have something which can be statically compiled, that is analysed and optimized in advance, just as much as something like C or C++. Without that, you need to either use an interpreter, or a JIT compiler, or a combination of both working together (like most current Javascript and some Python systems do, oh and like Java does even though it's not dynamic).

If course it's those dynamic features which attract developers because dynamic features let you write less code, writing code takes time, and time is money. It's all a trade off, and the best tool for the job depends on the type of project and the market. That's why we have so many programming languages.

The way that Asm.js is meant to be used is that you take libraries written in something like C++, and you compile them using their compiler to Asm.js code. Games is the big market being looked at, although no doubt there are loads of other applications. It won't replace run of the mill Javascript code though, just the stuff that has to be highly optimized for CPU performace.

The real competitor to Asm.js is Google's NaCl, which also uses an intermediate representation, although in their case its special NaCl code rather than a subset of Javascript. Looking at market acceptance though, it looks like Asm.js has pretty much won that battle, and NaCl is a dead end. It's also one of the final nails in the coffin for Flash and Silverlight.

Mozilla to whack HTTP sites with feature-ban stick

Re: why, why, why... what is the point?

@Russell Hancock - "In the last week I have looked at all the following, please someone explain why they need to be secure.."

Most of those sites you listed will be going to http/2 anyway, which is encrypted by default and has no unencrypted mode. They'll be going to it because it will give them better performance, use less bandwidth, and work better with mobile. Tests so far have shown http/2 with encryption to be faster than regular http without encryption. Chrome and Firefox already support it, and the other browser vendors will be following suit if they haven't already.

Mozilla's proposal is for what to do about sites that don't change to http/2 because they don't want to change anything. Their proposal is that those sites will continue to work as is for the foreseeable future, they just won't be able to access new browser features. Since those site operators who claim they don't want to change anything won't be accessing new features anyway, then they've really nothing to complain about, do they?

The only people who will be affected are those who want to use the latest bleeding edge web technologies, but don't want to do it over http/2 or encrypted http.

Re: Eeejits!

@Will Godfrey - "My own website is a very simple affair. No javascript and no flash, just a bit of css and some ordinary links."

Well it won't be affected by this proposal then, will it?

Re: Action. Counteraction.

@FF22 - Firefox relegates web sites that do not use HTTPS. Users relegate Firefox to the also-ran category.

Planning on browsing the web with telnet then? Google has already announced their own plans to achieve the same result. The other vendors will do the same if they haven't announced plans already. If anything, Mozilla are taking a much softer line on this than Google are. They're not working alone on this, as companies such as Cisco and Akamai are in it with them. The IETF, IAB, and W3C want some sort of solution, and even the US government is pushing vendors to come with something.

Mozilla will be making a proposal to the W3C. The browser vendors and various other interests will kick the idea around and come up with a common plan and schedule so that site owners will know what they need to do. Under Mozilla's proposal, existing sites will continue to work as is. It's when they want to access new features (e.g. getting access to your web cam) that they will need to do so through encrypted means (Firefox already requires this for some features).

The very first question in Mozilla's FAQ is "Q. Does this mean my unencrypted site will stop working?" Their answer is "Not for a long time" (they're talking to other companies about a joint plan for what to do over the long term).

So oh mighty ruler of the Internet, it appears you're panicking for no reason.

Ubuntu to shutter year-old clock unlock bug

Re: Shortsighted reaction on the side of Canonical

The discussion evolved quite a bit from his original description, which was rather vague and impractical. The person who reported it was working on finding a way to brute force an exploit, but at the time of writing this comment he still hadn't.

As for the original method, the Canonical guy said "If you have administrative users that are leaving session unlocked, you have a more serious security issue than being able to change the time." To make a long story short, if you have administrator access you can do administrative things. The Canonical guy isn't blowing the issue off, he's just saying 'show me a feasible exploit that doesn't require initial conditions that are not a serious security problem on their own'.

I'm not going to criticize the guy who reported this as a bug. I would rather have an overly paranoid set-up than an overly lax one. Even if he can't make a practical exploit out of it, someone else may find a way to do so later.

I won't be surprised if Ubuntu closes this by updating to the latest version of "sudo", but I have to agree that it's not a serious issue at this time. If you are worried about it, just use "-k" with sudo or close your admin terminals before leaving your computer logged in, unlocked, and unattended.

Microsoft HoloLens or Hollow Lens? El Reg stares down cyber-specs' code

Virtual Practicality?

So it's "virtual practicality" then? It doesn't really exist in our world, and it only works in the imaginary world of a Microsoft marketroid's presentation? Why this is unprecedented!

I expect to see this follow the trajectory of the MS Kinect - loads of hype, followed by public disappointment with the reality of it, and then disappearance into oblivion as only the most die-hard fanbois maintain any interest in it.

Visual Studio running on OS X and Linux for free? SO close

Re: some thoughts...

@Phil dude - "I am guessing there is substantial liability attached to planes, vehicles, industrial robots etc....?"

I've done a lot of work on industrial software (and hardware), and components like robots are part of a system in a larger machine or manufacturing cell. You design the machine on the assumption that software and hardware has bugs and that something like a robot could go wonky. If the operator can get hurt due to something the robot did, then that's the fault of the person who designed the overall system.

There are safety components such as light curtains, two-hand controls, safety relays, etc. which are designed to be "fail safe". These tend to be fairly simple things with very limited feature sets. You use these as building blocks to provide safety systems for use with for example your robot. Most complex industrial machines are one-off custom built jobs, so there's no universal "solution" to safety. The real problems must be solved at the individual application level.

At one time a popular brand of light curtain (used to disable a machine if you pass your hand through the curtain of light) used a software based system that used two different programs running in parallel on two different microprocessors (I think one was a Motorola, and the other an Zilog or Intel). Their later generations evolved to use ASICs to eliminate the software, and I think that everyone else did the same.

Safety shutdown systems in something like an oil refinery or a large boiler control are another much more complicated story altogether, and I won't go into that.

Feature set?

Any idea what this thing does that is significantly different from say Geany? The feature set in the documentation looks pretty run of the mill and I don't see anything worth the effort of downloading it and trying it out.

I know that programmers can get really attached to whatever editor they've been using for a while, but I've switched editors many times and there's not a lot of difference between most of them these days, unless you are talking about something really hard core like Emacs.

Intel's wristjob envy sparks reorg

Too Late

Would this be the same Intel who failed to make a dent in the mobile market? I can't see them having much success in this even more low cost and low power market. Small, cheap, and fast to market isn't in their DNA.

The companies that are going to amount to anything in this "wearables" market (assuming it isn't just another fad) are already tinkering with designs on the test bench. Intel is going to be "too little, too late" to make a difference.

Intel's big problem is going to be the relentless downward pressure on prices in a world where the CPU is a mutl-vendor commodity. It's the same problem that Microsoft faces with operating systems and it's why neither company has had much success in the mobile market. Microsoft at least has had some success at branching out into other software segments and so perhaps has a future after going through an IBM style painful restructuring. Intel however really has only overpriced CPUs used in stagnant traditional markets. So far despite thrashing about, there has been no sign of any ability to re-invent themselves as something different.

Microsoft is BEATING Amazon's cloud revenues. Er, how?

I can think of lots of well known third party companies who run their operations entirely on Amazon. I can't think of any who do the same with Microsoft Azure.

I tried googling for something about this Aaron Rakers from Stifel to see what he actually said about MS Azure, and all I can seem to find is older Register reports which last year said that Rakers claimed that MS Azure was beating Amazon, and obscure blogs which said that the Register reported that (you get the picture). Of this mysterious report itself, there seems to be no sign (I notice that it isn't linked to in the story). Nobody else in the IT press seemed willing to touch the story (including the earlier versions) with a barge pole.

So, I'm going to say that I'm a bit sceptical about the whole thing unless I see some real evidence. I also notice that the Reg journo seems to be holding the report out at arm's length like a long dead fish. I hope he got a decent lunch out of the Microsoft PR flack last Friday because otherwise his time was thoroughly wasted.

What is the REAL value of your precious, precious data?

Let's see, what has the US government been up to since they terrified their citizens into acquiescing the establishment of a police state to "protect them from terrorists"? Kidnapping people and imprisoning them in secret prisons. Torture of innocent people. Holding people for many years without trial. Need I go on?

The American security services buy data from commercial vendors. So tell me then, why would I want information about me to be sold to a country like that? Maybe, perhaps even likely, nothing would happen. But why take the risk? I mean, look at all the security theatre we go through to avoid the miniscule individual risk of terrorism. Why should we take the threat posed by the US Homeland Security any less seriously than we do terrorists?

If the American businesses who are complaining about this don't like it, then perhaps they ought to be spending their lobbying efforts (and money) on getting the American PATRIOT Act repealed? I suspect an awful lot of ordinary Americans would be pretty happy about that as well for their own sake.

Apple Watch shipments: Pick a number, double it. Hey, it worked for them

Rather Expensive

Phones typically have an 18-month replacement cycle, while watches stay in daily use for decades and even generations.

I suspect the intention is that every buyer of an Apple watch will have to buy a new watch to go with their new phone if they want all the features to work, and to not be more than one generation behind if they want it to work at all. This effectively doubles the price of an already expensive phone. There's undoubtedly a market for these things, but I don't think it's anywhere near the size of the market for just the phones.

This is the problem with the current "smart watch" concept. It's basically an expensive accessory for an expensive phone. When I buy a watch, I expect it to last at least 10 to 15 years, and I buy pretty cheap watches. Now find me a company in the computer or mobile phone business whose products remain 100% compatible for those sorts of time scales.

What it might do though is to undermine the demand for Apple's phones. If the things turn out to be really useful (and I'm not convinced they are), most people will find them unaffordable (witness Apple's long term shrinking market share). If people really want a smart watch to go with their phone, they may simply buy a cheaper phone (e.g. a Chinese Android) and a watch to go with it rather than pay a massive increase in their phone contract to pay for it. After all, if you're not pulling your phone out of your pocket and fondling it as often to see if someone sent you a new FaceTwat message, then who cares what it looks like?

Personally though, I expect the market will be limited once the novelty wears off. The screen on a watch is too small to do much more than display the time, and a regular watch does that rather well already.

Sysadmins, patch now: HTTP 'pings of death' are spewing across web to kill Windows servers

Re: Raise your hand...

@Mike Bell - Re: IIS market share "Nope."

You forgot to use the "joke" icon. Your own link shows IIS market share falling steadily down to third place since 2008, from 20 - 30% down to 10 - 12% for real web sites ("top million" sites are real sites, rather than placeholder pages held by domain speculators). The "top million" sites shows a steady smooth decline over time. IIS has been in third or fourth place for several years and is well on its way to oblivion. That's according to the data in your own link.

Seriously, I think that Microsoft knows that IIS is well on its way to being an ex-parrot and doesn't think there's much to be done about it. As a result, they'll give it basic maintenance and security patches, but they aren't taking it seriously enough to pro-actively get rid of the obviously bad ideas like the one which is the cause of this current security flap.

Since IIS is a legacy product, that's not too surprising. A lot of other vendors tend to take the same attitude to product lines which are on their way out to pasture.

Re: Raise your hand...

Well, there used to be an HTTP server available for Linux as an optional third party kernel module called "Tux". It let Linux win the web server "speed wars" willy waving contest with Windows and IIS which were (and still are) doing something similar. Then some time around the beginning of the century people realized what an incredibly stupid idea that was, so it never made it into the mainline Linux kernel as an official feature and it died from lack of interest many years ago.

However, it did get developers thinking about how to get data from the network port and back out again as quickly as possible. They came up with new software mechanisms which let user space HTTP servers work as fast as kernel based ones. In the end, there was really no valid technical reason for putting an HTTP server (or part of one) into the OS kernel. It was just used to play games with bench marks, and even there "proper" design ended up being just as fast as doing it in the kernel.

I'm surprised that Microsoft are still doing this. However, IIS is slowly dying off altogether so it probably isn't getting the sort of attention that something like this needs.

Nvidia's GTX 900 cards lock out open-source Linux devs yet again

Re: One of the reasons I abandoned them years ago

I also have to say that the AMD APUs have been great for running Linux for me. I'm using one to post this from Ubuntu 14.04 with Unity. It's using the Gallium open source video driver which comes with Ubuntu, so I don't have to fiddle about with proprietary drivers.

Overall, it's been very solid, fast, and trouble free with all the usual GUI stuff, including full screen video. For a Linux GUI desktop, I really wouldn't consider anything else at this time.

SQL Server 2005 end of life is coming, run to the hills...

@Trevor_Pott - "Just get them to port to Postgres."

I listened to a podcast interview with one of the principle developers of Postgres a couple of years ago, and he said that the largest number of users porting software from another database to Postgres came from MS SQL Server. This came as a surprise to the interviewer, as Postgres is generally considered to be more similar to Oracle than to MS SQL Server. However, that was their experience.

I take this as meaning that the differences between the details of how different databases work isn't that significant an issue when it comes to porting effort in many cases. I suspect this is because a very large proportion of applications don't really demand that much out of their database and don't use any of the more esoteric features. That is probably especially true for the SMB market that you mention.

Operation Redstone: Microsoft preps double Windows update in 2016

Re: LTS?

@keithpeter - " Ubuntu: LTS every 2 years each with 5 years support. Fedora: Basically acts as the input/testing platform for RHEL ..."

This is a problem the article has when comparing Windows to Linux. The Ubuntu and RHEL/Fediora models are utterly different from each other and the author doesn't understand them. There is only one Ubuntu (GUI and CPU choices aside), there's no such thing as a "community/freetard edition". What you download from Canonical is the same whether you pay them or not. Every six months they put out a development release, but every two years they put out an LTS. You can skip an LTS release and still have support.

Red Hat on the other hand handles things very differently. Fedora is generally packed with bleeding edge stuff, and has a support life of only a little over a year (the length varies). You get a month or two of grace after the next release, but then you have to upgrade if you want security fixes. RHEL is only available to paying customers (yes CENTOS is a RHEL clone, but it's not called RHEL) and it comes out at erratic, infrequent intervals like Windows does. It's not a timed release.

With Ubuntu, you upgrade in-place by clicking on a button (or by typing a command if you are using a GUI-less server). You upgrade RHEL by nuking and installing from scratch. Fedora attempts an in-place upgrade, but more often than not it fails. Most Fedora users don't seem to even try in-place upgrades anymore.

Ubuntu and Red Hat release on completely different principles. Comparing Windows to Ubuntu and Fedora tells us nothing about what Microsoft intends to do. Will it be like Ubuntu? If so, will there be non-LTS development releases? Will it be like RHEL - nuke and re-install? Or will it be like Fedora - upgrade every year, with in-place upgrades failing left and right?

Perhaps the author could have thrown in Arch just to complete the confusion - upgrade every day to the very latest off the fingertips of the developers, with things working most of the time, but with the occasional complete failure.

I suspect that the Microsoft Windows developers want something like Arch, sales and marketing want something like RHEL, and the IT press think that the result will be something like Ubuntu.

Choc Factory's king codec serves 25 BEELLION Tube hours

Re: Must be my eyes

In the H264 version, a lot of the transverse ceiling tile support frames are blurred out of existence, while they are visible in the VP9 version. Also, if you look at the woman in the middle who is split between the images, in the H264 version you can see jagged lines along the edges of her leg, while the edges on the VP9 version are clear. The same is true for straight edges along the walls as well. The "jaggies" are there to some extent in VP9, but they are much smaller and less prominent than in the H264 version.

The VP9 version definitely has a better picture. On the other hand, I'm not sure I would notice any difference if I wasn't actually looking for one.

I suspect the biggest advantage will be in terms of reduced bandwidth needs for VP9. That matters to YouTube who are paying a lot of bandwidth bills, and also to end users who are watching the videos via their mobile phone or capped Internet connections.

Since Google owns the dominant mobile phone OS and a major computer web browser, they can push the VP9 codec out to most of the market. Firefox also supports it, and it's built into a lot of newer televisions. On the source side, YouTube is a very big chunk of total Internet bandwidth. So if Google wants to use it, it's going to get used on a really big scale.

Samsung's bend blame blast: We DEMAND a Galaxy S6 Edge do-over

Glass is not meant to be bendy

I'm pretty sure that bending anything made out of glass is not a good idea. The bend tests done in the lab don't duplicate real life very well, so just because the glass on a particular phone didn't break in that test doesn't really mean anything. You could bend it a slightly different way and the glass may shatter.

Sticking a big (and expensive) phone in your trouser pocket is inherently not "safe". I wouldn't think it was a good idea to walk around with a glass drinking glass in my pocket and sit on it, so why would would I think that a big sheet of display glass would be indestructible or safe?

Really, I have to wonder whether any of these large smart phones are actually fit for purpose as "mobile phones". If you want to carry a phone in your pocket and not have to worry about it breaking, one of the cheaper models is a safer bet. They tend to be smaller, thicker (and so inherently stronger), and the really cheap ones have a sheet of plastic as the window over the screen. And if you do break it, you're not going to cry over the money you lost.

Nutanix looking for a way to burst VMware's bubble

Re: Docker

@K - Yes Docker tries to do one thing well, but that's its strength. It's not meant to solve every problem, but it solves a lot of common ones. Any complex IT system would consist of a mix of containers and traditional VMs, and indeed containers inside of traditional VMs.

As for what Nutanix is going to do, I agree it's more likely they are adding their own management tools on top of KVM than actually rolling their own hypervisor. The same may be true for Docker containers.

Pumping billions into data centres won't guarantee you an empire

Network Effects

I don't think that economies of scale involved in operating a data centre will explain the dominance of a few cloud providers. Rather it will be the network effects of all the third party software, developers, consultants, and services that will be built around the proprietary features of those clouds. A challenger would have to reproduce all of those in order to effectively challenge an incumbent.

It's like how Microsoft is dominant on the desktop, and Android is dominant in mobile. Apple can be a distant runner up in either of those spaces, but anybody else is a marginal player. It's the third party effects though which are the sources of that dominance.

Cloud providers will concentrate on increasing the network effects, and making it difficult for customers to leave once they're inside. They will also work on creaming off as much of the potential profits for themselves as possible, including by carving out promising lines of business to be exclusively for their own products.

If you believe that a free market is a good thing when it comes to economic performance, then this is a dystopian future. Economic growth will be weighed down by sub-optimal allocation of capital and poor use of resources. All in all, this is not a good thing.

Locally Integrated Menus back on Vivid Vervet’s menu

Re: Menus

@keithpeter - I'm using Ubuntu Unity 14.04, and it works fine.

"Steps to reproduce: Load LibreOffice Writer, click in the new document window, press and hold the Alt key while pressing the I then the O then the F keys then release the Alt key"

Try releasing the Alt key after pressing the "I" but before pressing the "O". You will notice that the underscore showing the top level key options will disappear when you let go of the Alt key. Holding the Alt key down lets you select a different top level menu option without having to cancel the operation and starting over. Once you let go of the Alt key point you can then select options within the selected drop-down menu by pressing additional keys on their own.

It's not "broken". This method is a lot more keyboard friendly if you're not entirely sure where the option you want is located. It lets you poke around in the menus to find something without having to use the mouse.

Re: Systemd

@Crazy Operations Guy - "I wish I was. I had an old box at work that was originally used with Vista Enterprise / Office 2007 (Company-standard Dell Optiplex GX 620)."

I suspect the CPU usage issue was a graphics driver issue. The newest versions of Unit and Gnome (as well as I think KDE) use hardware accelerated compositing, but they emulate this in software using the CPU if the graphics system doesn't support it. XFCE just doesn't do composiiting. Ubuntu has an XFCE version that probably would have worked quite well.

It's a trade-off, since on newer hardware it's much faster and more efficient to hand off the work to the graphics system, and even the very cheapest modern graphics chips (even the AMD APUs) support this. If you have one of the few low end older GPUs which don't, then the work gets thrown back on the CPU. However, even a very cheap $50 modern video card would be more than adequate for even the most demanding desktop (I'm not counting games though).

People using Windows went through this issue a few versions back when Microsoft also introduced hardware accelerated compositing. A lot of those people though just ended up throwing out their PC and buying a new one when the upgrade didn't work.

As for RAM usage, that's very difficult to compare between operating systems unless you are really intimately familiar with how each one uses and reports on RAM. Buffers are held and re-used by various systems and applications but may be released on demand. Getting repeatable results when one of the applications is a web browser is particularly difficult, because of the way they tend to buffer and cache complex pages after they have been processed.

Microsoft to slash price of top-level MSDN subs for Visual Studio 2015

Re: Seems a bit pricey

Admit it. You downloaded your copy of Vim off of the Pirate Bay.

Prostrate yourself before the GNU, commands Indian DEITY

Re: Would The Reg please stop

@Christian Berger - " I used to work in a large appliance manufacturer with a "no FOSS" policy. This is because one of the owner companies was sued for violating the GPL because they didn't respond to a complain within 2 months or so."

Was that an IT appliance manufacturer, or a kitchen appliance manufacturer? If the latter, they must really have struggled to find a way to violate the license. Even for the former they must really have worked at it because the number of actual FOSS copyright lawsuits that I have heard of can probably be counted on the fingers of one hand (mainly Busybox in embedded systems), and those have always taken years of ignored lawyers letters from the copyright holders before they took the step of going to court over it.

And I have to wonder just how sympathetic Microsoft or Oracle would be to pirating their software if you simply ignore their complaints. Those companies sue loads of people all the time over license violations.

Anyway, their IT setup must be interesting, because they can't use Java since that's open source (GPL) and loads of Java libraries and frameworks are only available as open source. Their heads must be ready to explode now that Dot Net is supposedly becoming open source.

If Microsoft drops out of the phone OS business, they are going to be rather limited in their choice of mobile phones since both Android, Apple, and Blackberry phones all have substantial FOSS bits that can't be removed.

Secret Bezos delivery helicopters operate from mystery Canadian base to evade US regulators

The CBC has a much better and more detailed article on this:

http://www.cbc.ca/news/technology/amazon-tests-delivery-drones-at-a-secret-site-in-canada-here-s-why-1.3015425?cmp=rss

As for the size of the packages, I believe they will deliver up to 2.3 kg. That probably covers a big chunk of what people buy.

However, I have serious doubts about whether there is much demand for it. Future delivery improvements are more likely to revolve around conventional truck delivery to unmanned clusters of pick-up boxes where you can go to pick up the item at your convenience instead of a delivery truck coming to your door. This will reduce costs while still giving reasonably quick delivery.

Day FOUR of the GitHub web assault: Activists point fingers at 'China's global censorship'

Baidu hasn't been hacked. This is supposedly happening somewhere else. Some part of the network (possibly the routers) is injecting Javascript into the response traffic after it has left Baidu but before it reaches the user to add extra Javascript. ISPs in the US have been doing the same thing to inject advertising, but in this case the Javascript conducts a DDOS instead of showing ads.

From the sounds of it, Baidu wouldn't even be able to see anything different from their data centres, since the Javascript injection happens somewhere else in the Internet, possibly on its way out of China.

I imagine that Baidu is not happy about this, since it would have the potential to hurt their business. The article isn't clear on this, but it describes the traffic being intercepted as being from "Baidu's advertising network", so it's quite possible that Baidu is losing a significant amount of money on this.

Smart meters are a ‘costly mistake’ that'll add BILLIONS to bills

Re: Smart, huh?

@John Lilburne - "The bulk of the known sources of uranium aka yellow cake is mined in Chad which is not exactly stable"

According to Wikipedia, the only mineral which Chad mines is small quantities of sodium carbonate. I've not seen any source which says they produce any uranium at all.

The number 2 uranium producer is Canada, and Australia is number 3 (Kazakhstan is number 1). I'm pretty that neither Canada nor Australia are about to fall over due to political instability.

Chrome trumps all comers in reported vulnerabilities

Re: I wonder how they measure vulnerabilities in Gentoo and Solaris.

I just had a look at Secunia's database, and it looks like they count vulnerabilities in all the software that can run on Gentoo against Gentoo itself. In other words, a Java vulnerability counts against Gentoo, and so do problems with Chrome, Flash, etc. I don't know how they end up with fewer vulns for Gentoo than for Chrome alone, but that might just be because Gentoo might lump several Chrome vulns into one notice.

Secunia is counting by vendor, and since Gentoo redistributes lots of third party software then all the third party stuff which could potentially get installed gets counted against Gentoo. On the other hand, the exact same software may have the exact same vulnerabilities on MS Windows or Apple OS/X, but it won't be counted against Microsoft or Apple because they didn't distribute it.

I think Secunia simply counts notices put out by vendors, they don't actually analyze them and apply any judgment. This means that the more conscientious and detailed a vendor (or distro) is about informing their customers (or users), the higher the vulnerability count they will have.

It also means that you can't actually compare vulnerability counts between vendors unless they operate, distribute, and report in a similar manner. That would just be comparing apples to oranges. I'm sure though that won't discourage our favourite anonymous security commentard from ignoring the facts and stuffing both feet in his gaping pie hole. Let me save him some trouble - "Microsoft had zero vulnerabilities while Linux had seventy-bazillion and caused global warming as well". There, now where do I pick up my cheque from Microsoft?

As to why Gentoo has loads more security vulnerabilities reported than any other distro, I suspect that is simply due to differing reporting and repo support policy. If another distro has smaller supported repos with fewer third party software packages, then they will pretty obviously have fewer vulnerabilities to report on to their users. Note though that I said supported repos. Different distros have different support policies.

I'm not sure what the story is for Solaris. I didn't bother looking them up in the database, and I'm not sure what their distribution, support, and reporting policies are. I wouldn't be surprised though if a lot of the apparently high vulnerability count is also simply due to double counting of non-Solaris related problems combined with a long support life.

Big Data high priest Stonebraker anointed with Turing Award

I listened to a podcast which interviewed him a couple of years ago. I can't remember the name of it unfortunately or I would link to it here. However, he came across as an eminently practical person who relied on hard measurements and real benchmarks instead of making assumptions and waving his hands about. He is a hugely influential figure in the modern software world through the work he has done, but his name isn't widely known because he wasn't someone to spend his time tooting his own horn or caring about getting ludicrously rich.

He's still doing very important work, involving things like SQL databases that are as fast as the popular No-SQL ones. He definitely deserves any and all awards given.

Snowden dump details Canadian spies running false flag ops online

Re: Wow!

The news story didn't quite quote it correctly. The number refers to items of network traffic that were believed to be associated with a "threat". In other words, emails, HTTP GETs, log-in attempts, etc. that looked unusually suspicious. The slides date from 2009, so I imagine the numbers are a lot bigger today.

The original source wasn't clear, but it appears that this number may be an aggregate across the entire "5 eyes" rather than just counting "threats" directed at Canada alone.

As for the "125GB of internet traffic per hour", that's counting just the "metadata", not the full network traffic that goes with it. I also imagine that they've increased these numbers considerably in the past half dozen years.

First figures in and it doesn't look good for new internet dot-words

Re: Is it wrong?

@I ain't Spartacus - "What would happen if someone manages to poison Google so that a search for hsbc gives hsbc.scam, instead of hsbc.com?"

So is "royal.bank" the Royal Bank of Canada (Canada's biggest bank), the Royal Bank of Scotland (weren't they a bit shaky?), or the Royal Bank of Nigeria (I just made that one up)? I guess I'll just have to google "Royal Bank" and click on the first link that comes up. Or I could just look at the business card the nice lady at the bank gave me when they set my account in the first place.

@I ain't Spartacus - "you weren't allowed to register a dot.bank unless you were regulated by a legitimate national central bank (or banking regulator)"

Because that works out so well for all the crooked banks in all the massively corrupt countries around the world. Do you have any idea how easy it is to get a financial institution license in a lot of countries? Do you have any idea how many "banks" are just brass plates or file folders in obscure third world countries?

@I ain't Spartacus - "the registrar could operate a national page"

Or the UK government could simply set up a ".bank.uk" sub-domain and tell the handful of existing UK banks to use that. Then it would be entirely under UK law in the same way the banks themselves were and you wouldn't have to worry about the honestly and integrity of the Royal Bank of Nigeria.

Re: Is it wrong?

@I ain't Spartacus: "The other thinig I can see working is the likes of dot.bank. If they genuinely only allow regulated banks in, have some proper standards, proper security"

You're going to give all your money to someone you've never heard of before just because they have a "dot bank" web site? Seriously?

"But they'd have to have some more secure way of finding the domain your after"

Or you could, oh, I don't know, perhaps just ask your bank?

Re: Telling quote

The more gTLDs there are, the more pointless any of the newer ones become. It's like ".biz" versus ".com". I just googled "dot biz", and the top two results were the Wikipedia article, and an SEO web site page explaining why having a "dot biz" is pointless if you want anyone to take your web site seriously.

Wind turbine blown away by control system vulnerability

Re: Sir

@Anonymous Coward - "Software is written by the same people with the same level of knowledge of software security"

I follow control engineering forums where this sort of question comes up. It's usually phrased as "the customer has just told us that he wants to check his wind turbine over the Internet. Is there some sort of web box thingy that I can add to the PLC to do this?" They then buy an eye-wateringly expensive "module" from the PLC vendor that lets you load a web page using MS Frontpage and associates a selection of ActiveX or Java "web controls" directly with PLC memory addresses. They're all designed to be installed by people who have never created a web page before and wouldn't know an HTML tag if it bit them. Difficult problems like oh, let's say security, are dealt with by not having any.

For 99.9% of people doing control engineering work, their knowledge of things Internet or web is limited to knowing where the good porn sites are. The control hardware they use is sold by companies whose knowledge of things webby is little better.

Guardian: 'Oil reserves will soon be worth NOTHING!' (A bit like their stock tips, really)

Re: Critical point missed.

An even more critical point is that the bulk of the world's oil and gas resources are not owned by the well known major oil multi-nationals. They're owned by governments who are more concerned about long term revenue streams than they are about stock market valuations. These are Saudi Arabia, Iran, Russia, Qatar, the UAE, etc. Most of these countries also have national oil companies who finance their investment out of cash flow, not the stock market.

The big western oil multi-nationals operate around the fringes of that. What those well known multi-nationals do or don't do won't make much difference on a global scale.

Zombie SCO shuffles back into court seeking IBM Linux cash

A few corrections.

"The case hinges on just who owns Unix"

- Not correct, that has already been settled years ago. Novell (now owned by Attachmate) owns Unix. SCOG was only a licensee, plus they also subcontracted as the agent to collect the license fees owed to Novell by the other licensees (e.g. IBM, Sun, etc.). The reason that SCOG declared bankruptcy was to freeze the case involving a suit with Novell (bankruptcy court overrides regular civil courts) days before SCOG was about to be handed its arse over breach of contract and walleting money that belonged to Novell. SCOG however was still solvent at that point. They then used bankruptcy status as a legal shield to let them do anything they wanted while their victims were prevented from retaliating. The execs and lawyers then ran through the assets of the company while protected from the shareholders, creditors, and anyone else who might have wanted to put a stop to the gravy train.

"animated only by the prospect of one day getting IBM to write it a cheque"

- IBM is also suing SCOG, see above for why this has been frozen. The IBM suit was the original one, but it got put on hold while SCOG sued Novell to try to force them to transfer ownership of Unix to SCOG (on the grounds that they couldn't sue world+dog without it).

SCOG's case involving Linux had already collapsed before that however, as they couldn't find anything they owned which had been copied into Linux.

SCOG's case with IBM revolves around the partnership their predecessor (the real SCO) had for a new version of proprietary Unix that was supposed to cover the full range of hardware from low end x86 up to large IBM servers. SCO was contributing their x86 expertise, and their marketing channel for the small business sector. IBM exercised their contractual right to back out of the partnership when SCO sold their Unix business to Caldera (who then renamed themselves the SCO Group - SCOG).

IBM saw the writing on the wall, and realized that Linux was going to fill the market niche that the proposed joint version of Unix was aimed at. The original SCO probably saw that as well, which is perhaps why they sold their Unix business.

The whole "unfair competition, etc." rubbish in SCOG's complaint has to do with IBM backing Linux instead of continuing with the Unix joint venture after SCO sold it to Caldera (who ironically were a Linux distro, similar to Red Hat or Suse).

SCOG sold off the remnants of their Unix business (SCO Open Server and SCO Unixware) to someone else a few years ago and have no real assets at this time. Their legal firm had signed up to handle the case to its conclusion in return for a lump sum (I think it was $30 million) plus a share of the promised loot, the prospect of which evapourated not long after (following which these same lawyers moaned that they were getting screwed).

Trustees are running SCOG, there's nothing left to fight over, and everyone wants the case to be wound up. However, the case can't get off the books until the lawsuits and counter-suits are settled. IBM didn't want to settle out of court because their reputation was at stake. They wanted to nail SCOG's hide to the wall. I won't be surprised if the trial is very short with SCOG's case getting tossed out, IBM's counter-suit winning, and SCOG's affairs being wound up and liquidated shortly after.

Cisco posts kit to empty houses to dodge NSA chop shops

Re: @Trevor Don't buy US kit

@Trevor - Winston Churchill supposedly once said something along the lines of "the Americans can always be counted on to to the right thing, once they have exhausted all the other possibilities". Right now they're still near the top of the list of alternatives, and it's a very long list.

This sort of story will put pressure on large American companies who will see their export markets melt away as IT technology become more and more vital and valuable. The fact that someone, somewhere else in the world may be worse isn't going to make customers trust them. It's like asking which of two paedophiles is the least bad one to babysit your children.

Ultimately the real answer to the "Cisco-NSA" problem is going to have to be technological. People were already thinking along the lines that we need better security, but the Snowden revelations have given that process a big kick in the pants to move the process along. If the NSA can hack into your systems, then so can anyone else, including common criminals. Either you're secure, or you're not. Right now most companies and individuals are storing their data in the equivalent to a cash box in a desk drawer with the key taped to the top of it.

Re: Don't buy US kit

@Trevor_Pott - Guess what story popped up in the Globe and Mail today?

NSA trying to map Rogers, RBC communications traffic, leak shows

http://www.theglobeandmail.com/news/national/nsa-trying-to-map-rogers-rbc-communications-traffic-leak-shows/article23491118/

The NSA has been hacking into RBC (Royal Bank of Canada - Canada's largest bank) and Rogers (one of Canada's largest communications companies). Oh, and into Rolls Royce (giant UK aero engine manufacturer) and Rio Tinto (giant UK mining company). And that's just a selection of companies beginning with "R". That's a pretty good cross section of major multi-nationals who compete with American companies.

Now I'm waiting for our usual AC commentard to tell us how it's vital for the US to hack into Canadian banks and British aerospace companies in order to protect the world from terrorism.

Re: seal? anyone?

The NSA was having the Cisco kit diverted to a location in I believe the Washington area, where they had a facility set up to open packages, install whatever it is they did, reseal the box, and send it on its way again. They apparently have copies of all the necessary Cisco seals and packing materials so you can't tell the difference. It's almost a small production line, not someone doing it at his desk at NSA headquarters. This was in the report which Cisco is trying to respond to in this story.