| 2004 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2005 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | |
| 2006 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Oct | Oct | |
| 2007 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | 2007 |
| 2008 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | 2008 |
| 2009 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | 2009 |
| 2010 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | 2010 |
| 2011 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | 2011 |
| 2012 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | 2012 |
| 2013 | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | 2013 |
Fri 31st Jul 2009 @ 23:23 2009: Please take your card out so we can read it
I am doing a fair amount of train travel at the moment, and Virgin Trains have a "FastTicket" system whereby you can book your tickets online, and pick them up from a machine at the train station, thereby skipping the queues for the ticket office (replacing them with the queues for the FastTicket machines, but I digress...)
The procedure is that you insert the credit card that you used for the booking ("something you have") and type in the alphanumeric code that you were given at the time of the booking ("something you know"). That is a relatively sane security policy, so fair enough (though really it would probably be good enough to verify that the booking is for "Mr Parker" and you hold a card in the name of "Mr Parker", bug - again - I digress).
This week, I put my card in the machine, but I was a little confused by this message on the screen (click for larger image):
Thu 30th Jul 2009 @ 19:05 2009: F1 Drivers' good wishes to Felippe Massa
http://www.youtube.com/watch?v=TXGBjwNyOMY - who knew that Kubica spoke Spanish?
Wed 29th Jul 2009 @ 00:06 2009: ESR - Eric Raymond - unstable
ESR quotes an eyewitness who by her own account saw nothing; ESR cites this in itself as "cause for war". ESR threatened Bruce Perens' life in 1999. It appears that since the CML2 debacle, that he has become increasingly unhinged.
Sad to lose a member of the community who contributed The Cathedral and the Bazaar - an excellent book, very much of its time, and capturing the essence of the F/OSS development model and community at the turn of the century.
Tue 28th Jul 2009 @ 16:34 2009: Understanding Sun in Three Easy Steps (1 of 4)
I think I've spotted what went wrong with Sun - Jonathan Schwartz can't count...
Sun 26th Jul 2009 @ 00:31 2009: OpenOffice - grr - Error 529
Err529
If I have understood this correctly, it seems that OpenOffice.org (OOo) can fall over when a bunch of normally-numeric cells contain an item of non-numeric data.
I have just gone through what is (to me, not a big spreadsheet user) a pretty large spreadsheet replacing all "looks blank to me" cells with a "zero", until the sums added up to the results that I was expecting.
Maybe this is an understandable issue; I am really unqualified to say. MS Excel and OO Calc both seem to act very strangely to my simple engineering mind. I did think that we generally accepted that 1+2+3=6.
If you have OOo, take the attached 123.ods spreadsheet (7.5Kb):
| One | Two | Three | Total |
|---|---|---|---|
| 1 | 2 | 3 | 6 |
If you replace the "2" with "x" (or even with " "), it seems to turn the cell into a "text" field, which makes OOo incapable of calculating the result, so the "Total" field (E3) becomes the inscrutable "Err:529".
The values are:
B2="One" C2="Three" D2="Four" E2="Sum"
B3="1" C3="2" D3="3" E3="=B3+C3+D3"
For some reason, the set of evidently invalid data that I had ended up putting into a spreadsheet meant that I was getting wildly inaccurate data from OO Calc. I prefer wildly wrong to subtly wrong, but my confidence in OOCalc for future use is reduced by this experience.
I have used OOo to create an XLS version of the spreadsheet (6.5Kb - interestingly, smaller than the OOCalc version); I would be interested to know how Excel deals with this situation. I do not have MS Excel to hand at the moment.
Build details (Debian "testing"): OpenOffice.org 3.0.1 OOO300m15 (Build: 9379) openoffice.org-core 1:3.0.1-9, Mon Mar 30 00:04:38 CEST 2009
Fri 24th Jul 2009 @ 23:58 2009: Maker's Schedule, Manager's Schedule
Paul Graham of YCombinator has written a new article, Maker's Schedule, Manager's Schedule, which seems like one of those things which is obvious when it is explicitly pointed out, but until you start to think of things in those terms, are far from obvious.
"The manager's schedule is for bosses. It's embodied in the traditional appointment book, with each day cut into one hour intervals. You can block off several hours for a single task if you need to, but by default you change what you're doing every hour.... But there's another way of using time that's common among people who make things, like programmers and writers. They generally prefer to use time in units of half a day at least. You can't write or program well in units of an hour. That's barely enough time to get started.
"Each type of schedule works fine by itself. Problems arise when they meet"
Thu 23rd Jul 2009 @ 22:22 2009: Swine Flu
Question Time tonight has mentioned the swineflu website being overloaded; a number of times it has been mentioned in the programme that this is because non-sick people are accessing the website.
This is absolute bull-flu. If the website needs to handle the load of concerned citizens and can't handle it, then the website is underpowered. It is not difficult or expensive to scale a website appropriately.
$ telnet direct.gov.uk 80
Trying 81.19.104.51...
Connected to xansa2.lbwa.verio.net.
Escape character is '^]'.
HEAD / HTTP/1.0
HTTP/1.0 302 Found
Date: Thu, 23 Jul 2009 22:17:42 GMT
Content-Type: text/html; charset=iso-8859-1
Server: Apache
Vary: Accept-Encoding
Location: http://site220.gslb.theclubuk.com/en/index.htm
Connection closed by foreign host.
Tue 21st Jul 2009 @ 23:45 2009: Must... Remember... Passwords!
One of the many great things about ssh is that it can use PKI for authentication, so you can put your public key on a remote server (here, it's called "somewhereelse.example.com") as ~/.ssh/authorized_keys, and log in without a password.
Well - I say "without a password" - you need the password for your private key (optionally, you can have a passwordless private key, but that would be very very very silly).
So you don't need to have the password on the remote server.
Thing is, I forgot the password, and the private key is on a machine two hops away....
Here is the (slightly anonymised) transcript:
steve@onebox:~$ ssh steve.example.com
Enter passphrase for key '/home/steve/.ssh/id_rsa': (enter) - onebox's key doesn't get me in here
steve@steve.example.com's password: (steve.example.com password)
You have new mail.
Last login: Mon Jul 20 18:40:28 2009 from irrelevant
steve@steve.example.com:~$ scp anotherbox:.ssh/id_rsa my_private_key
steve@anotherbox's password: (anotherbox password)
id_rsa 100% 1743 1.7KB/s 00:00
steve@steve.example.com:~$ exit
Connection to steve.example.com closed.
steve@onebox:~$ scp steve.example.com:my_private_key .ssh/anotherbox_id_rsa
Enter passphrase for key '/home/steve/.ssh/id_rsa': (enter) - onebox's key doesn't get me in here
steve@steve.example.com's password: (steve.example.com password)
my_private_key 100% 1743 1.7KB/s 00:00
steve@onebox:~$ cat .ssh/anotherbox_id_rsa >> .ssh/id_rsa
steve@onebox:~$ rm .ssh/anotherbox_id_rsa
steve@onebox:~$ ssh steve.example.com
Enter passphrase for key '/home/steve/.ssh/id_rsa': (enter) - onebox's key doesn't get me in here
steve@steve.example.com's password: (steve.example.com password)
You have new mail.
Last login: Tue Jul 21 20:45:29 2009 from onebox
steve@steve.example.com:~$ rm my_private_key
steve@steve.example.com:~$ exit
Connection to steve.example.com closed.
steve@onebox:~$ ssh-add
Enter passphrase for /home/steve/.ssh/id_rsa: (id_rsa password)
Identity added: /home/steve/.ssh/id_rsa (/home/steve/.ssh/id_rsa)
steve@onebox:~$ ssh myotherid@somewhereelse.example.com
Last login: Mon Jul 20 13:10:17 2009 from irrelevant
myotherid@somewhereelse.example.com$ passwd
Changing local password for myotherid.
Old password:
Damn! I still can't remember it!
So I am stuck with the same password on myotherid@somewhereelse.example.com, because I can't remember the old one!
Must speak nicely with
root or at least one of the wheel oligarchy!Tue 21st Jul 2009 @ 11:29 2009: 32-bit NetBackup client
If you get this message from NetBackup (eg, bp, bprestore, bplist, bpbackup, etc):
EXIT STATUS 100: system error occurred while processing user command
Then run it through strace:
# strace -o /tmp/strace.txt ./bprestore -B -C myclient -R /tmp/change.txt /etc/hosts
and see if it is loading 64-bit libraries for a 32-bit binary:
# grep lib64 /tmp/strace.txt |grep -v ENOENT
Make sure that you have 32-bit versions of these libraries and that they are earlier in LD_LIBRARY_PATH than the 64-bit versions:
# export LD_LIBRARY_PATH=/mnt/runtime/lib:/mnt/runtime/usr/lib:$LD_LIBRARY_PATH
Specifically, this is when using RHEL5.3 with a customised stage2.img which includes a 32-bit NetBackup client, even though the hardware is 64-bit.
Fri 17th Jul 2009 @ 00:23 2009: IE6 Troll
This is a response to http://carbide20.com/2009/06/06/ie-splash-pages/. I will not make this a link, as the images are NSFW (Not Safe For Work). They are basically a tirade of offensive images which the website suggests could be provided instead of actual content, to users of Internet Explorer 6. No code is provided; presumably similar abuse is also intended for those who are stuck with even older versions of Internet Explorer.
Anyway... here is my response, in case anybody cares:
Yes, IE6 is a dinosaur, and its continued existence makes life hard for web developers.
But many people are forced to use it by corporate standard desktop builds.
Most users who have the choice, have moved to a newer IE, or to a better browser (FireFox, Opera, anything but IE). Most users who do not know better, have been forcibly upgraded by Microsoft patches.
So the only remaining IE6 users are those who are forced to use it because their standard, corporate desktop includes IE6.
Ironically, this is often because it would cost too much money to pay some web-monkeys to upgrade the intranet sites to support newer web browsers.So it's actually the crappy HTML that web developers wrote years ago which means that the crappy web browser is still alive.]
I am currently on a contract at a place which mandates IE6, and the web proxy actively blocks the installation of any other browser. Presumably because the intranet requires IE6. Because too many web developers were too lazy to bother with web standards when they could make easy money by claiming "so long as it works in IE6, which has a massive market share, those obscure web browsers shouting about web standards don't matter."
So it is fascinating to see those same trolls now launching a backlash against IE6, and suddenly in favour of standards, after they've made an entire industry out of websites documenting how to make stuff work in IE6 as well as with standards-compliant web browsers.
Thu 16th Jul 2009 @ 23:54 2009: Three website -often down. But why so chilly?
Silly HTTP/500 response pages:
Why is a chirpily cold blonde woman a suitable picture to represent "sorry, our website is down again"?
Still, it's better than Barclays can manage for their Priority Club branded credit card. When the website does respond, the best it can manage is this:
This is worse again: "New Document"? Please, at least try to pretend that you made an effort...
Thu 16th Jul 2009 @ 23:39 2009: Sun Stockholders approve Oracle takeover
Press Release
Sun and Oracle generic web page
Thu 16th Jul 2009 @ 19:20 2009: Upgrade for -£3.00?
Well, for the sake of -£3.00, I might as well have the upgrade!
Fri 10th Jul 2009 @ 15:13 2009: Wonderful disclaimer
The Linux kernel, on booting, can be passed a parameter "vga=xxx" to tell it about the capabilities of your display adapter. It means that it can do pretty displays as it boots up. The documentation (/usr/src/linux/Documentation/svga.txt) includes this wonderful disclaimer:
If you are not happy with the mode list offered (e.g., if you think your card
is able to do more), you can enter "scan" instead of item number / mode ID. The
program will try to ask the BIOS for all possible video mode numbers and test
what happens then. The screen will be probably flashing wildly for some time and
strange noises will be heard from inside the monitor and so on and then, really
all consistent video modes supported by your BIOS will appear (plus maybe some
`ghost modes'). If you are afraid this could damage your monitor, don't use this
function.
I love F/OSS for the cavalier attitude to disclaimers!
Tue 7th Jul 2009 @ 00:46 2009: Mono and the GPL
There has been a lot of discussion lately about Mono (a Free Software implementation of Microsoft's .NET platform) and the GNU General Public License (which is the license behind most of the software that makes up a GNU/Linux distro).
The Mono project have been working for a good few years now, to make a .NET compatible platform for GNU/Linux systems, just as .NET works on MS Windows systems for the language they call "C#". How successful they have been at this, I am not in a position to judge; I don't write .NET code and I don't use any .NET software.
As many people may be aware, the Microsoft way of distributing software, and the GNU/Linux way (particularly the Debian GNU/Linux way) of packaging and distributing software, are quite different.
Microsoft's approach may be brutally summed up as "We wrote it, we own it. You may pay us for the privilege of using it."
The Debian GNU/Linux approach may be similarly butchered as "It was written under a Free Software license; you get it on the same terms that we got it - we (and you) can take it, use it, change it, redistribute it; but it can't be redistributed it under different terms than these."
The biggest difference here, is the redistribution. You can't resell a modified Windows; you can resell a modified GNU/Linux. So what is in a GNU/Linux distro matters. So when it was suggested that the next Debian default (and full) install would include an Mono app called Tomboy (which has a non-Mono equivalent, GNote), Richard Stallman (amongst others) stated his concern about the potential legal (patent) issues involved. Stallman says, "If we lose the use of C#, we will lose them [other apps written in C#] too. That doesn't make them unethical, but it means that writing them and using them is taking a gratuitous risk."
There has been some silly mud-slinging since, which is best ignored. Today, though, this link has been brought to the attention of the GNU/Linux community: http://www.microsoft.com/interop/cp/default.mspx. It claims a publication date of September 2007, with an Updated date of Feb 2009. This link has been cited by some pro-Mono people as a response to the anti-Mono coverage.
However, it does not seem to provide any clear assurances;
Q: What if I don’t implement the entire specification? Will I still get the protections under the CP?
A: The CP applies only if the implementation conforms fully to required portions of the specification. Partial implementations are not covered.
That is a long-winded way of saying "No".
Q: Does this CP apply to all versions of the specification, including future revisions?
A: The Community Promise applies to all existing versions of the specifications designated on the public list posted at /interop/cp/, unless otherwise noted with respect to a particular specification.
Or - in other words: "No.".
Q: Is the Community Promise intended to apply to open source developers and users of open source developed software?
A: Yes. The CP applies directly to all persons or entities that make, use, sell, offer for sale, imports and/or distributes an implementation of a Covered Specification. It is intended to enable open source implementations. Because open source software licenses can vary you may want to consult with your legal counsel to understand your particular legal environment.
It is not clear to me (IANAL) what the definition of a "Covered Specification" is.
Q: Is this Promise consistent with open source licensing, namely the GPL? And can anyone implement the specifications without any concerns about Microsoft patents?
A: The Community Promise is a simple and clear way to assure that the broadest audience of developers and customers working with commercial or open source software can implement the covered specifications. We leave it to those implementing these technologies to understand the legal environments in which they operate. This includes people operating in a GPL environment. Because the General Public License (GPL) is not universally interpreted the same way by everyone, we can’t give anyone a legal opinion about how our language relates to the GPL or other OSS licenses, but based on feedback from the open source community we believe that a broad audience of developers can implement the specifications.
Bit of anti-GPL FUD there - the GPL is clear enough, thank you. This is a question about the Microsoft license, not the GPL license.
Q: I am a developer/distributor/user of software that is licensed under the GPL, does the Community Promise apply to me?
A: Absolutely, yes. The CP applies to developers, distributors, and users of Covered Implementations without regard to the development model that created such implementations, or the type of copyright licenses under which they are distributed, or the business model of distributors/implementers. The CP provides the assurance that Microsoft will not assert its Necessary Claims against anyone who make, use, sell, offer for sale, import, or distribute any Covered Implementation under any type of development or distribution model, including the GPL. As stated in the CP, the only time Microsoft can withdraw its promise against a specific person or company for a specific Covered Specification is if that person or company brings (or voluntarily participates in) a patent infringement lawsuit against Microsoft regarding Microsoft’s implementation of the same Covered Specification. This type of “suspension” clause is common industry practice."
Hang on - so it does "absolutely, yes" cover me, or not? I think that the answer here is "No."
As a Debian GNU/Linux user, I do not want, will not have, and maintain the freedom to not have, Mono as part of my OS environment.
As a GNU/Linux professional, I very strongly suggest Sun's Java implementation over any others, for compatibility reasons - even having multiple JREs installed in different JAVA_HOMEs for apps that require a particular environment. I can not see a need in the Enterprise for a copy of Microsoft's .NET platform, which can (unless I have seriously misunderstood the situation) only play catch-up with Microsoft's implementation.
Mono; non-Sun Java - why have them at all? If you want .NET, why not use Microsoft Windows?
