phone transcript
"911. What's your emergency?"
"Someone bought Microsoft Office instead of Open Office!"
"Sir, this line is for real emergencies only."
Posts by Don Mitchell
316 posts • joined Sunday 29th April 2007 18:34 GMT
Jim Gray
Jim didn't work on DEC's servers, he was a research scientist known for expertise in data base systems. One of the best technical writers ever, his book on Transaction Processing is one of the bibles of database design and theory. Jim worked at Tandom, IBM, DEC and Microsoft.
One of Jim's projects was Terraserver, the forerunner of Google Earth. Hosting complete satellite image coverage, his group had to buy military spy satellite images from Russia. He also worked with astronomers to develop a unifying database of sky survey images, which was only completed after his death (the World Wide Telescope).
Jim was a great guy, and a great scientist, and his disappearance at sea caused a lot of sadness. Ironically, there was a large effort to search for his boat using online satellite images.
Posted in Amazon takes on Microsoft Azure head on
EC2 vs Azure
Amazon is dominating the cloud computing market. One place they cannot compete with Microsoft is in developing their own sophisticated software services, like MS Dynamics. And of course they are tying Azure services in with Win8.
Azure itself is pretty cool OS technology. Culter's team spent a year just studying the formal requirements for geographically distributed data centers. They merged technology from NT, Hypervisor and (I'm guessing) the SQLOS layer. The scalability tests I've seen are surprising, way better than Linux and even better than NT at multicore parallel performance.
What makes me skeptical about Microsoft is their executive management. I still would not be surprised if they fail. Their lasting impact may be the disporia of former Microsoft systems engineers, who are scattered throughout the industry today (for example, running EC2 at amazon). Systems engineering rules. You don't learn it in college, and it's not part of the open source culture, so if you want to built something big and complex, you gotta find those industrial experts.
Posted in Amazon takes on Microsoft Azure head on
Re: Azure vs AWS
You don't know what you're talking about.
IBM once had 70% hardware market share
What is this new devilry? Something from Microsoft?
No. It is IBM, a demon of the ancient world.
This foe is beyond any of you...run!
Posted in Battlefield Earth ruled worst film EVER
Famous for being bad
BFE isn't really the worst movie or even the worst sci-fi movie. But it has become the most famous-for-being-bad movie, which is why it always wins these contests. Same thing just happened on io9. Still...it is kinda bad.
Why do we care what Greenpeace thinks?
Greenpeace: misanthropic, anti-growth, anti-technology, dogmatic, puritanical, zealous, upper-middle-class-white, romanitic pasturalists.
Posted in Teens break up with Facebook
Re: Anecdotal
I think Google+ has different data-mining advantages. Facebook gets information you explicitly decide to enter by clicking Like. Google gathers data by monitoring your activity, what youtube videos you watch, browser search strings, URL strings (if you use Chrome), scraping gmail text, etc. It's probably more complex for Google to analyze its data, but it has the advantage that much of their data is aquired involuntarily, from all users regardless of what they "like".
Posted in Vote now for the WORST movie EVER
Battlefield Earth meme
The problem is that Battlefield Earth is so famous for being bad, it always wins thsee polls, even if it is nowhere near the worst movie on the list.
ebook
It's well over ten years since Microsoft cancelled its ebook tablet project. Trying to unfumble the future now?
Posted in Why Windows 8 server is a game-changer
Re: seems a bit arse about face to me
Linux is not a natural choice for desktops, in fact it has failed enormously in that area. The compeition is all on the server side. If you talk to people at big ecommerce sites, it's a choice between stable software and big fees from Microsoft, or much less stable Linux which has lower direct fees but lots of associated costs of hiring programmers and dealing with frequent emergencies.
At one major ecommece site in Seattle, they call their Linux programmers "ferel engineers", because of the total lack of any kind of systems engineering ethos in the OSS hacker community. They depend on a handful of senior engineers to herd cats and maintain some order in the process of keeping a complex website up and running 24/7.
One of these engineers described Linux as a single-app OS -- you set up your website and then tune Linux until it doesn't crash anymore, and then you don't even breath on the system for fear of destabilizing it again.
Re: "Windows is dead."
Exactly. People who use office are not going to abandon keyboards and big screens and do all their work on an iPad. And it's not a zero sum game, people own multiple devices. Microsoft is just going to sell more software and add value to the iPad.
Posted in Amazon floats free Windows Server clouds
Thunder in the clouds
It will be interesting to watch Amazon and Microsoft battle over pricing. Microsoft has the most sophisticated OS technology, with Cutler's Red Dog system. But Amazon has done a better job of building and managing cost efficient data centers.
I worked at Bell Labs during the 1980s. BSD wasn't really viewed as proper UNIX by the purest, although our lab ran it instead of 32V (the research version of UNIX). It fixed a lot of problems, basically by incorporating VMS features into UNIX, but was considered too bloated and badly programmed. Last time I talked to one of those guys, I asked him about Linux, and he thought it was more bloated and badly documented than Windows. The idea of keeping an operating system clean and simple is long dead...
Roo: Total Ignorance about OS History
I'm impressed that you know Dave Cutler's name, given your lack of knowledge about operating systems or history. When NT was written in 1989, it had dozens of modern OS features that UNIX either did not have or was still in a very confused state. Cutler's impact on UNIX is what is not properly appreciated. I was at Bell Labs when UC Berkeley folks ported BDS to the VAX, and it was pretty clear to us that BSD was UNIX + VMS (virtual memory, a file system that was not journally but at least did not horribly suck like UNIX V7).
Linux is basically UNIX + NT. So many ideas in modern UNIX come from Microsoft - the use of dynamic linked libraries, device driver interfaces, asynchronous file I/O, journaling file system, etc. All things done by MS operating systems before UNIX. UNIX still lacks the systems engineering design that Cutler and Microsoft brought to operating systems, the modularization and formal interface (e.g. COM) structure. Using BSD in the last 1980s, we had to edit tables and recompile the whole kernel to install a new device driver, since drivers were simply subroutines in a monolithic program.
Readers who are interested should take a look at Hart's book on Win32 Programming, to see what a kernel design should really look like.
10 watts/square meter
I look up acres and annual kilowatt hours on these plants. This new site gets 10 watts/m**2, which is not bad for photovoltaic panels. BUt solar-thermal plants are getting 30.
The First Cloud OS
The first cloud OS that isn't Dave Cutler's "Red Dog" system.
Russian Mars Missions
Fourth attempt? Mars-1 to Mars-7, Fobos-1 and 2, Mars-96 and quite a few others that were stranded in parking orbit and given some "Kosmos" or "Sputnik" designation. A few others, like the two M-69 orbiters, never made it to orbit.
Mars-5 was essentially successful, it went into orbit in 1973 and carried out all its mission tasks (photography and numerous spectroscopic and space-plasma and cosmic-ray experiments). Fobos-2 entered orbit and performed some of its tasks, but failed before it could rendezvous with Phobos.
Let's not forget that Russia's Soyuz/Fregat technology sent Mars Express to its destination. Mars Express consists of a communications satellite refitted with the backup copies of Mars-96's experiments. So there is considerable Russian involvement of the "pay no attention to that man behind the curtain" style.
Fobos-Grunt failed to achieve three-axis orientation lock on the Sun and a bright star. Those sensors are redundant, so this may be a software glitch. It will be in range of special telemetry and telecommand stations in Baikonur Wednesday night, and they may be able to get a full set of housekeeping telemetry then. There is a small chance the mission can be resumed, if this is just a fixable software state.
E-Books
I'm still waiting to hear why MS killed their E-Book project in the late 1990s. They had something like a kindle, invented ClearType as part of that project, but then just dumped the project.
Ritchie > Stallman
One of the reasons Ritchie is not more famous is because of Stallman's hatred for the Bell Labs group and the practical suppression of pre-Torvalds UNIX history. Remember, GNU stood for "GNU is not UNIX". They hated UNIX/C, because it came out of a corporate research lab, and most of all because eclipsed the TOPS-10/LISP culture that the MIT AI Lab dominated in those days.
Yes, part of UNIX success was that it was free and open source. But it was also elegantly simple. I worked at Bell Labs Research in the 1980s and knew Dennis and Ken. I heard horror stories about MULTICs, how it was so complex and inefficient it could only "time share" two users at once. Ken came off that project determined to do something completely different. Simplicity is hard to achieve, and a lot of thought went into UNIX and its (now long-lost) philosophy of composing simple tools to perform complex tasks.
There was a long rivalry between MIT and Bell Labs about this question of design simplicity. The hacker culture had a macho attitude of "my code is bigger than your code", while Ken and Dennis spent hours trying to boil things down to the fewest lines of code and the fewest necessary features.
I remember one attempt to reconcile. The UNIX team invited Stallman to visit Bell Labs. I don't recall much about Stallman's talk, but everyone remembers that he picked his nose and ate a bugger in front of everyone. In return, Rob Pike was sent to give a talk at MIT, which he was not able to deliver, because Stallman and his friends heckled him.
Something New Now?
Maybe instead of reverse engineering Microsoft Office, someone will actually create something original that deserves to survive in the market place.
Posted in Google+ chases MySpace for second place
Visitors != Users
Most people I know have made a Google+ account, visited it a few times, and then never returned.
Google and Facebook are the two giants in the internet advertising space. I'm not really cheering on one of them to eat the other.
Why do computers crash
I build my own computers and install Windows without any of the usual bloatware from SONY or Dell. I also build Xeon based machines using ECC memory. On my lastest machine, running Windows 7, I haven't seen a single crash. I think there are two problems that cause some systems to crash. One is memory errors from non-ECC memory, a growing problem with gigabytes of high density memory. One cosmic ray boring its way through some 32 nm DRAM will flip a lot of bits! There was an internal study at Microsoft some years ago that found that 50% of blue screens were caused by hardware memory parity errors. THe other major cause was device driver bugs, a frustration that Microsoft has gotten some control over in Win7 by sandboxing the drivers (I think they run in in ring 1 instead of ring 0, but I could be wrong about that).
The other reason some people seem to have sick PCs is more nebulous. The typical user downloads a lot of crapware, never defrags their drive, and is careless about security, and at some point they seem to end up with a machine best treated by "Format C:" and starting over.
Reasonable
Sounds like a reasonable choice. Amazon certainly has the best service and expertise. Azure yet to prove itself on as big a scale, but the underlying new OS ("red dog") is an exciting project. Probably Dave Cutler's last opus.
Old Work
Microsoft has been working on this stuff at least as long as Apple. MS Surface, developed by Dave Kurlander, was one of the very first multi-touch user interfaces. You can blame MS for not deploying this work and taking more risks, but it's naive to just say they stole the ideas from Apple. Gavin hates Microsoft, we all get that, really hilarious. But it would be nice to get some objective reporting now and then.,
Posted in IBM PC daddy: 'The PC era is over'
IBM, UBM, We all BM for IBM
IBM sold their PC business a long time ago, and view the PC and Microsoft as an enemy now. An IBM vice president once said he wanted to put an ice pick into Bill Gate's head. So not really interesting to hear them say the PC is dead.
We can be thankful that IBM created the PC and then blunded by letting it become an open platform that they couldn't control.
Who knows what the future of personal computing is? A device with a big screen and a keyboard is not going away any time soon. It's clear that consumers value a mix of products (TV, cell phone, desktop computer, etc).
Forget Apple and Android
Buy a Samsung Series 9 laptop. It's thin and made of black anodized duralumin, if you drop it 8 inches, it won't shatter into white plastic pieces like a Macbook. Load Windows 7, MS Office, Photoshop, Visual Studio, ABBYY FineReader and a few hot 3D games.
SQL
You can also get free versions of MS SQL Server. Dig around and find TPC benchmarks for MySQL and you will see what a joke it is for transaction processing. It's especially true on multi-core hardware where the professional database systems are way beyond the open source toys.
It also begs the question of why you would be using Apple systems to do any kind of serious server work. That's for NT or Linux.
MS database
MS systems come with SQL Server, which is much higher performance than MySQL.
Is there secrecy on the internet?
He who LOLZ last LOLZ best. Ultimately, the packets run over corporate satellites and AT&T photonic switches and undersea cables. Almost every packet on the internet goes through NSA filters. So, it may seem like there is a hidden virtual world of hackers, but I think that will turn out to be an illusion. The actions of Anonymous and LulzSec just encourage the owners of the internet to develope better ways to track our activity.
No quicktime please
I hope we will not have to install quicktime or itunes to use hulu. still, it could be worse. Google is also talking about buying hulu ... you're watching hulu, and hulu is watching you!
eBooks & Surface
I worked with the eBooks team in the late 1990s, so I saw Microsoft's vision for tablets. The problem was not a lack of imagination by the developes or product teams, it was a lack of imagination by Ballmer and his executive vice presidents. The only cool idea to survive that project was ClearType.
They also failed to leverage the MS Surface tecnology, invented by researcher Dave Kurlander. You see that multitouch UI in the new versions of Windows, and everyone assumes they just copied the iPhone, but in fact they had this a long time ago and just failed to make good use of it.
Posted in Can Big Blue survive another century?
Ugh!
"A demon of the ancient world. This foe is beyond any of you. Run!"
I used IBM systems in the 1970s and 1980s, and they were absolutely horrible. Job control language was arcane and complicated and reflected a totally wrong model of files and processes. They had a hideous time sharing system which MIT helped them design, basically consisted of typing virtual card decks of JCL and then submitting them as batch jobs. Early versions of UNIX came with a "TSO Shell" that mocked it. Next time someone tells you MIT invented timesharing, bonk them on the head with a printout of Dartmouth BASIC (which actually did have a positive impact on early timesharing technology).
IBM was innovative in the 1950s, but small inventive companies like CDC surpassed them in the 1960s. Then came DEC and mini computers, then workstations based on single-board systems like SUN, and finally the PC. It's lucky for us that IBM's PC was designed with off-the-shelf parts that anyone else could build. And of course they blew it by letting Microsoft own the OS, which they sold nonexclusively to everyone. IBM hates Microsoft with a captain-ahab passion even today.
And let's not fool ourselves about IBM supporting open source. They are an enemy of Microsoft, and like every other organization that cannot write an operating system themselves, they ported UNIX. *gulf clap* But don't hold your breath waiting to see the source code to DB2.
A much needed reform in patent law has been lost
A lot of computer companies were backing Microsoft on this case, because it involved a much needed change to patent law. it's too bad that in the gleeful bashing of Microsoft, that is being overlooked. This has been an important case in IP law, a lot more important than the arguments of Linux vs. Windows fanboys. This would have made it easier for everyone to invalidate bad patents, and the system if choked with patents of obvious ideas and already-published ideas. This ruling against Microsoft was not good news.
Yes Please!
I know a lot of Microsoft employees who agree, Balmer has been a disaster since the bumbling of the antitrust trial in the 1990s.
Microsoft still has a lot of talent and some great technology (SQL Server, .NET CLR, the NT Kernel, Office, Xbox, etc), but it has become top heavy with unimaginative executives, with Balmer being the most obvious. What happened to the eBooks project in 2000? Why did they fumble the phone and pad markets when they had done a lot of fundamental research in those areas? They just failed to execute on new ideas, because their management was too afraid of risk.
They might still have time to recover, because outside of Apple and Oracle, they don't have many high quality competators. Linux is still a kludging unstable mess that costs companies a fortune in maintanence. MySQL is still a third the speed of SQL Server and Oracle, and it doesn't scale up with parallel hardware very well. The software market is Microsoft's to lose, and Ballmer is the man to lose it.
The UNIX community didn't grok Windows for a long time
At Princeton prior to 1995, we had a variety of UNIX workstations in the CS department, IRIX, HPUX, a lab full of SUN's for the undergraduates, and if you were really low on the pecking order, you got an X-terminal that logged into a DEC mainframe running UNIX. If you looked at anyone's screen, what you saw was half a dozen command-shell windows running various text-oriented programs. It was still very much a command-line interface world, still heavily dependant on text editors to do any kind of content creation. That was the state of the art in UNIX. You could copy/paste text between windows, but there was none of the object/embedding/scripting action that you had in Apple or Windows PC. You were just beginning to see people play with TCL/TK, an ersatz version of Visual Basic, but UNIX "apps" were still very limited by the lack of underlying system support.
At a deeper level, UNIX was still very monolithic in those days, while Windows was very modular and object oriented. UNIX was just beginning to support DLL's and device driver interfaces, so the operating system was pretty much one giant C program. Windows was based on dynamically loaded libraries and COM interfaces that let Microsoft update and swap out components without compiling from source. I still remember the pain of installing a new scanner on my SUN workstation, having to fiddle interrupt vector tables and then compile the whole damn kernel. This lack of modularity plagued the UNIX window system as well. You can go read the X-windows paper in ACM ToG, and it's all about how clever they were at supporting overlapping of bitmaps and handling redraw with you moved something. But there's nothing there about the underlying logic of communicating interface objects that Apple and Microsoft were focused on. A UNIX window was a virtual command console. If a program presented a GUI interface, it was pure brute force, it had to track the cursor and know where any "buttons" were.
Even in the late 1990s, when SGI briefly flirted with Windows NT on their hardware, it was pretty amusing to watch them demonstrating drag and drop, like it was something new discovered on Mars.
JPEG2000
I've wondered about why JPEG2000 hasn't really taken off. It may be partly patent encumberment and partly that it is just too slow and costly to encode/decode. Microsoft developed a faster encoding that is similar to JPEG 2000, but I don't know the details of it.
WebP uses DCT block coding, like jpeg and mpeg. It almost doesn't pay to get too clever about "better" encoding, because DCT has been optimized so intensely, to the point of influencing processor instruction architecture.
Not even remotely "revolutionary"
Timesharing. X-Termainals. Network Computers. Thin clients. We've seen all this before, and we know it sucks. How is this not strictly less powerful than a laptop PC, which can run a browser and do thousands of other apps and games and not be a useless brick when there is no network connectivity?
Data Center Failures
And you think if you roll your own data center, it will be more reliable and cost effective?
Biased reporting
Because they dislike Microsoft, and are continually attempting to imply it is technologically inferior.
The greatest disappointment I have in the open source movement is that it didn't create a new state of the art OS, but instead just ported tired obsolete UNIX to the PC. NT is showing its age too, but at least in 1990, it was state of the art, which cannot be said for any point in the history of UNIX.
Beating a dead horse
Come on guys, have you ever reported anything positive about Microsoft. You clearly 1) hate this company like lots of people do, and 2) don't really know that much more than anyone else about them. Why bother?
Posted in Windows Server pushed to the super limit
Amdahl's Law
Operating systems start to level out in performance growth due to shared resources. For Linux, this starts to happen around 4 cores, for Windows at about 8. Microsoft has been doing multiprocessor server software long before Linux even was modified to do multiprocessor. NT learned its tricks on expensive server hardware built by UNISYS, back in the early 1990s, well before modern multicore processors were common.
Sophisticated appliations like SQL Server scale up to 128 cores, because they are designed to use the OS in special ways and because there is a special scalable layer called SequalOS that handles a lot of systems activity without even calling on NT.
Dave Cutler's new "Red Dog" operating system has been rumored to scale up to 128 cores do to extensive kernel redesign.
Politics
Now we will see who has the best lobbyists and political connections. Microsoft got the shaft a decade ago, because they didn't give enough money to congressmen and influential academics. I think they are still behind Google. Look at the dinner Obama had a few weeks ago in Silicon Valley, the political connections and influence of Kleiner Perkins and friends is enormous.
Finally
I have to wonder why Microsoft killed its eBook table project ten years ago, or why they didn't capitalize more on Kurlander's pioneering work on touch interfaces for the "Surface" project (basically all the concepts of the iPhone on a table-top display). It's not that they lack the talent or brain power, they lack the guts at the executive level to take risks.
How to build a super computer
So to summerize this article, buy lots of rack-mounted servers, and install Linux and Appache, Wow. And here I thought the relevant problem was the AI software.
Cloud vs. Pseudo-Cloud
Real cloud computing is a great idea, letting people rent high quality data center resources in a very flexible way. That market is dominated by Amazon now, and Azure will be a big player too. Amazon and Microsoft are empowering users and entrepreneurs to build new web services.
But what Google means by "cloud" is something else, vastly less empowering. They are talking about owning our data and owning important high level services like document processing. It's a totally self-serving definition of "cloud", and it's coupled with their thin-client operating system.
PDP-11
We used a PDP-11/34 and later an 11/70 at Caltech's Space Radiation Lab. The 11/60 let you write user microcode. Not a bad machine, but just no market for that capablity.
I don't agree with the comment about PDP-10 vs. VAX. But certainly the PDP-10 had a huge influence. A whole generation of academics grew up on TOPS-10, and there was quite a culture clash when they met up with Bell Labs UNIX. The modern UNIX is sort of a blend of philsophies, Bell Labs ideas about simple tools and modularity with the TOPS-10 style of big feature-rich command tools. "ls | c" vs. "ls -c". Not that much simplicity is left in modern bloated UNIX systems.
VMS was a good system, not the greatest user interface, but much better technology under the hood than UNIX. Berkeley tried hard to bring UNIX up to the level of VMS, with a decent file system and virtual memory and proper libraries and linkers. BSD never could beat VMS's performance, and UNIX never had really good compilers like DEC's until pretty recently.
The alpha was not too bad, but a huge power hog. To get clock syncronization, they drove a separate metal layer with a meter-long transister (folded up into a little square in the center of the chip). "TICK! TICK! TICK!" was what people jokingly said about alpha's clock.
The lessons
The lessons from this seem to be 1) never install anyone's browser toolbar and 2) don't run Chrome. If you don't want your info being uploaded. I stopped using Chrome when it uploaded info to Google that caused it to merge some of my accounts that I wanted to keep separate. It also stuck one of my email addresses on my blogger page then, and not the correct email. Google just works too had at gathering and correlating information about users.
Posted in Gates: Killing the internet is easy
Disinformation
@Tom Welsh, that's complete disinformation that you are simply making up.. I saw the internal memos from Bill Gates and Nathan Myhrvold in 1994 and 1995, and they were extremely excited about how they would change the internet by flooding it with Windows users. Futhermore, MSN was never a closed proprietary network, it gave open access to the internet from day one. That actually forced AOL to change its policy and open up more. And who even rememers compuserve or prodigy, which tried to remain closed networks.
History lesson
IBM trying to come back is like unearthing a Balrog, some ancient evil from the distant past. We can thank IBM for one thing, being naive enough to make the PC an open hardware platform. They seem to still hate and blame Microsoft for the fact that they never really cashed in on the PC, but in retrospect it is a good thing they did not do as Apple did and make it a close platform. Ultimately, even Apple had to adopt it.
Popular Whitepapers
- The BI Inflexion Point
Information is a right, not a privilege - VPN security - if you want it, come and get it
Attention WiFi hotspotters: You want it - The Register Guide to iSCSI
A primer on Internet SCSI, a protocol to transport SCSI commands over IP - Secure Mobile Working
Beyond the Technology - The Impact of IT Security Attitudes
Putting the pieces in place for effective security delivery - The Register guide to unified communications
A primer on the implications of unified communications for enterprise IT
