Bryan Lawrence : Bryan's Blog 2005/06

Bryan Lawrence

... personal wiki, blog and notes

Bryan's Blog 2005/06


I'm pleased to say that after a silence due to

  1. a significant ankle injury that meant that there was no comfortable way to approach my computer for significant lengths of time, and

  2. upgrading the site to leonardo-0.6.1

... i'm back!!

Upgrade notes here.

2005/06/30 (permalink)

Terms and Conditions

eliterate librarian has drawn my attention to the LexisNexis website and their terms and conditions, which are a joke.

The implication that I can be bound by terms and conditions which I have not read (although actually I have in this case, but am ignoring), is as, I say, a joke. Our legal experts tell us that if you want a disclaimer to hold, it has to be obvious. Here is what my browser shows me of the LexisNexis web site is this:

Image: static/2005/06/30/LexisNexis.jpg

the point being that unless I scroll down, I wont see the terms and conditions link, and even if I do, would I assume I have to read it before I could link to any part of their website? Very unlikely. The terms and conditions that eliterate librarian highlights, and you can go laugh at, are useless ... (at least in UK law ... while it is a dangerous assumption, I assume US law is just as sensible about holding people to contracts they haven't seen)!

by Bryan Lawrence : 2005/06/30 : Categories badc curation (permalink)

Hurricanes and Climate Change

It has been an interesting six months for followers of the issue of whether hurricane intensities and frequencies are changing in response to changes in the background climate.

This has been a bit of a cause celebre, with some claiming that trying to make the link is evidence of duplicitous intent by "the climate scientists". Of course, the key problem is that not every part of the climate system is exhibiting signals of climate change, nor may they necessarily do so. However, it's obviously appropriate to ask the question, but regrettably having done so, the media charges on and marks such links as explicit and authoratitive well in advance of scientific opinion. Crichton in his book at least had that part right - although it's a bit much to imply that governments are provoking it ...

Anyway, as I say, an interesting six months. It began with an impassioned open letter from a respected hurricane scientist, Chris Landsea, who thought that Kevin Trenberth had gone too far in linking climate change to the intensity and frequency of hurricane landfall in the U.S. (particularly in later press conferences).

Roger Pielke wrote a measured followup note explaining the necessity of a peer reviewed paper explaining Trenberth's position.

A paper along those lines has now appeared. Trenberth's point, which I think is fair is that a) there is a plausible physical mechanism which could link global warming to more intense and frequent thunderstorms, but b) at the moment the observational evidence is still not unequivocal (especially in terms of statistical tests).

With respect to his mechanism, it's really rather simple:

Both higher sea surface temperatures and increased water vapor tend to increase the energy available for atmospheric convection, such as thunderstorms, and for the development of tropical cyclones.

... and there is observational evidence for both of those. There is also considerable simulation work 1 that demonstrates that this mechanism could work in practice, but not yet 2 unambiguous evidence of hurricane landfall and intensity changes.

1: See this doi: 10.1175/1520-0442(2004)017<3477:IOCWOS>2.0.CO;2 - I can't link to it directly because of the inconsiderate use of <> in the doi, which I'll have to think about handling (ret).
2: By using the word yet, I admit to an opinion about what I expect to happen as the record gets longer and the carbon dioxide builds up, but it's just that, an opinion! (ret).

by Bryan Lawrence : 2005/06/17 : Categories climate (permalink)

Laptop OS Woes

Well, my dicky stomach has lasted pretty much all week. While one major side affect has been the inability to concentrate on anything, that has allowed me to set the computer off doing things in the corner while I watched the clouds (or yesterday, the patches of blue on the clouds) ...

So, I did some real testing on kubuntu suspend, which broke! Broke so badly that it caused a kernel panic and corrupted the root partition ... so that's the end of kubuntu for me for now! (Despite what I said on Sunday).

Meanwhile, back in the Suse corner, I did get it to install finally. All it took in the end was to install with the safe settings. I suspect the dma between the dvd and the hard disk ... Couldn't Suse have suggested that in the dozen useless emails we exchanged? Apparently not.

I'm now back in the horrible land of Yast. I hate not getting any (useful) feedback on software installation progress, and in particular, what to do when it hangs, and abort fails! (Which seems to be most times).

The astute reader will recall that back in April I suggested I could be a candidate for a Mac. But even that option seems silly now, given that Mac laptops are due to move to intel next year. (I appreciate the Osborne Effect isn't a necessary way to behave, but given I might actually consider dual booting mac and linux, at least during a transition phase, it makes sense to wait).

by Bryan Lawrence : 2005/06/16 : Categories computing (permalink)

Subversion and Roundup

A while ago I was investigating choices for issue tracking and code maintenance, and was bemoaning the difficulties of trac (which sounded good, but was just too hard to get going).

Some interesting things popped up in todays reading (I'm banished to the spare room - the one with the computer in it - because I've got a dicky tummy, and so I can't do anything "useful") ... this means I'm reading deeper into my feed list than I have for a while.

Anyway, two things of interest:

by Bryan Lawrence : 2005/06/12 : Categories computing ndg (permalink)

kubuntu 5.04 wins over Suse 9.3

Well, I think I'll give up on Suse. I tried to install it in another partition on my laptop today (so I could follow the advice from the support guy), but this time it didn't even install the software onto disk properly ... (but it installs fine onto my desktop at home). It got 20 minutes into the software install, informed me that a component had failed, gave me the option to retry or ignore. Retry did nothing. Ignore hung the system into an infinite loop.

So, it appears that kbuntu has won for now. I've started to put notes on kbuntu here!

Update (16th of June): Oh no it hasn't (won). See on why not.

by Bryan Lawrence : 2005/06/12 : Categories computing (permalink)

More on Software Patents

A couple of key happenings:

  • On the one side, we have Tim Bray summarising a legal argument by Greg Aharonian to show that the Microsoft XML (office) patent will never be asserted because of copious prior art.

  • On the other, Groklaw are attempting to draft suitable wording for a definition of "technical contribution" for purposes of EU patent law.

These represent respectively, the silliness that exists, and the silliness that could exist ...

by Bryan Lawrence : 2005/06/05 : Categories msxml curation (permalink)

MS Office Licensing

I think it's great to see the new blog from Brian Jones at Microsoft (via Sam Ruby). Brian has a specific entry on the licensing question for Office documents, but as he says, he's not an expert on that area. Some of the comments to that entry hit the buttton though:

Microsoft has lots of high-powered and smart lawyers. They know very well that this license is incompatible with GPL'ed software, like, for example, OpenOffice.

Again, I think the analysis on Groklaw is the best I've seen on this, and again, until this is satisfactorily resolved, I have a big problem with relying on ms documents to store my own IPR for posterity (i.e. beyond the next couple of years).

by Bryan Lawrence : 2005/06/03 : Categories msxml curation (permalink)

Playing with Kubuntu and Korganiser

As regular readers will know, I'm a disappointed Suse Linux bloke at the moment. I've had a support ticket running for a fortnight now, but haven't made much progress (to be fair to Suse, i've been slow in replying to a couple of emails, but to be fair to me, they never read properly what I write, we haven't actually moved past the information I provided in my first email yet).

Anyway, in a fit of frustration because I want a few things now, I installed kubuntu while I was having a bath last weekend (it took me all of a few minutes to set up the software download, a few more to burn a CD, a few to start of the installation, then I had my bath, and when I got out, I had a few minutes config to do). Well done kubuntu for that.

Kubuntu is a bit rough around the edges ... there are a lot of things that don't work quite right1... it's not quite as good an experience out of the box as Suse normally is, but at least it installed!! I've found playing with apt-get much better than yast ...

This afternoon I had a spare hour, and instead of doing something I should do, I've had a play with kontact, korganiser, and the calendering functionality. I'm starting from being a regular kmail user who uses outlook via cross-over office for calendaring from our corporate exchange server. I have two choices from here. I move over to evolution, or I hope that korganiser can support exchange properly ...

Starting from the latter because I like kmail, I've had a play with korganiser and the exchange 2000 plugin (in kde 3.4.1). It's very nearly functional :-). I have my appointments visible, and can create new ones, but I can't yet from this interface see who is coming to the meetings, and/or their free/busy status for new meetings. Still it's certainly progress from the last time I looked.

1: if I persist with it, I'll document them ... (ret).

by Bryan Lawrence : 2005/06/03 : Categories computing (permalink)

Adobe slightly better than Microsoft

I'm on record with my complaints about how Microsoft license their document formats. Imagine how aghast I was when I read this - apparently the Adobe Acrobat license conditions are stupid. Go read it for yourself ...

... and then step back and realise these are the conditions to use their software! The data in the documents are still safe (or are they?).

It's fine for companies to protect their software IPR, but not the products that we create with their software. That's our IPR. How can it be that even smart Microsoft employees (i.e. See Dare Obasanjo in this and this) can't understand why this makes those of use with responsibility for preserving information for posterity very worried? Tim Bray has it right again.

by Bryan Lawrence : 2005/06/01 : Categories msxml curation (permalink)

DISCLAIMER: This is a personal blog. Nothing written here reflects an official opinion of my employer or any funding agency.