readingtype

Firefox goes for Meta. Is there nothing better?

2010-03-14T12:14:00.005Z

Good to see the Firefox 3.6 ‘what’s new’ page is swathed in styles that use Erik Spiekermann’s font Meta. But a shame that Mozilla or its design team couldn’t find an OFL licensed font worthy of their use.

A better OFLB online today might have helped. The design and coding is done; we await implentation.

Screen scraping joy

2010-03-09T10:14:00.005Z

Been doing a touch of screen scraping, scripting with Ruby, against a target that was ‘unwilling’. A few observations:

Using Mechanize (available in various forms for Perl, Python and Ruby [homepage for the latter]) is a must. I started with the Ruby HTTP library, then went to Curb (Ruby’s implementation of Curl), but having the pages you retrieve abstracted into an object that you can manipulate in familiar terms (like, say page.forms_with :name => "choose_colour") helps you concentrate on the peculiarities of your task
Replicating the path of a real user is important. There could be session variables at the server end that mean jumping about between items that you cannot navigate between as a regular user will generate error pages, but see below
Don’t count on friendly HTTP errors from the server, as it might not know it has done anything wrong
If the page output looks OK but you cannot parse it, check for funny characters hidden in the HTML. I found ASCII nulls dotted about; these are initially hard to spot for somewhat obvious reasons. Browsers can deal with this kind of dodginess but XML parsers, as @fidothe reminds me, must ignore the elements in which such characters occur. I was able to do this to get around the problem:


@agent = Mechanize.new
class << @agent
  alias :orig_get :get
  alias :orig_fetch_page :fetch_page
  # remove the chaff characters
  def get(options, parameters = [], referer = nil)
    page = orig_get(options, parameters, referer)
    page.body = page.body.gsub(/"[0x00]"/,"")
    page
  end
  def fetch_page(params)
    page = orig_fetch_page(params)
    page.body = page.body.gsub(/"[0x00]"/, "")
    page
  end
end

[0x00] represents ascii null in the sample code; I was able to select and paste the character from an HTML dump with both vim and a GUI text editor but it tends to be less than visible in the wild and YMMV.

Assume that what you’re doing is an unwelcome task. If the points above don’t give you that impression, other curiosities probably will.

SERIAC 2010 in Chertsey

2010-01-31T10:39:00.003Z

I’ve attended SERIAC (the South Eastern Region Industrial Archaeology Conference) for the last three years and enjoyed it every time. It’s also very good value at £12.50 for a day of talks that invariably span the whole field of industrial archaeology but remain understandable and interesting. This year it’s hosted by Surrey Industrial History Group at Chertsey Hall in Chertsey on Saturday 24 April. The programme is online; all the talks look genuinely interesting. Hope to see you there.

Subsections in Textpattern

2010-01-25T10:57:00.004Z

One provocative omission from the excellent Textpattern CMS is subsections. I don’t know why it was omitted (probably just because it was out of scope when Dean Allen sat down to write TXP). Now that there’s a codebase, fitting in subsections is slightly tricky. There’s a need to adjust the way that URL rewriting works and there’s also a need to rework the section admin form. I could really do with having this sorted and I have started to scope it out. If somebody would like to sponsor the work let me know ’cos I haven’t got time to do any more for nothing.

Yes, I have investigated what’s out there. It is not good enough, I’m afraid: for one thing, it is susceptable to break with every minor point release. Something better is needed.

Web fonts talk online

2010-01-19T11:08:00.004Z

The text and the slides for the talk I gave last night at London Web Standards are online.

In the talk I gave suggestions about how to get started with web fonts which are still a bit of a mess. I’ve also cited a lot of very helpful information on the net so you can get much more detail on the topics I covered. I’ve also stuck the html, css and js test files I used on there. Comments and corrections most welcome.

Running 14k, sterling

2010-01-04T16:13:00.001Z

Just need to get this off my chest.

I’m self employed, work pretty hard: no holidays, no luxury, no TV. I do have a car but I walk, cycle or train it when I can. I keep the heating low and shop in the most meagre fashion. I’ve got two pairs of presentable trousers. It’s not just long habit, this miserly behaviour, it’s become necessity — don’t ask me how or why but I always thought I was doing the right thing recording income and expenditure and meekly handing the numbers to HMRC.

So a tax bill for £14,076 hasn't made me feel particularly well loved. It’s about half of a really good year’s net income for me (and about three quarters of this year’s). Suffice to say it’s far, far beyond what I can afford.

The moral of this is don't do your own tax return unless you know what you're doing. Regardless of whether I can crawl out of this one, the ongoing corrosion of tax has cramped my style for far too long. It hasn't made me happy, it hasn't made me productive, it's just given me grey hairs.

Thats all, cheers.

Call/response on climate monitoring station records an opportunity for open data?

2009-12-06T15:43:00.004Z

Nature’s editorial of 2 December called for various measures to make it easier for climate scientists to produce the analyses that are used to model the world’s climate. These underpin the current thinking on the probability and effects of a warmer climate in the near future, and the essential ingredient is data. According to the editorial the data is subject to international agreements on its publication. So it is good to see an announcement from the Met Office on 5 December to the effect that data from 5000 temperature monitor stations should be available from ‘early next week’. The agreement has been influenced by the World Meteorological Organisation, the UN’s weather body.

To a layman, the two obvious questions are how accessible, and how useful, this ‘subset of the full HadCRUT record of global temperatures’ will be.

The Digital Economy Bill: a letter to Lord Puttnam

2009-12-01T12:00:00.005Z

Dear Lord Puttnam,

In response to the proposed Digital Economy Bill, on 25 November you said:

'I also add a broad welcome to the proposals that throw a spanner in the proliferation of online piracy.' [1]

I agree that legislation is appropriate in an area that is forming an increasingly significant part of our working and social lives.

I also strongly support the concept and practice of copyright as a pillar of the economy, although I reject the somewhat natural urge to extend the terms. I support copyright because I know that it gives me the ability to allow people to benefit from my work under the terms of permissive licences (thus ultimately benefiting myself), and because I recognise that unregulated copying will undermine a creator's ability to support themselves -- which ultimately means that our society in its current economic form will not be able to create new and original work in any field.

But I feel that to describe the proposals for curbing piracy [the distribution of material subject to copyright without permission] as 'throwing a spanner' is about the best that can be said for the accuracy and effect of the proposed Bill. From what I have read it seems as though there will be very damaging outcomes from the application of its clumsy provisions, but ones which may not harm the offenders at all.

Perhaps your statement tacitly acknowledges that the Bill in question needs much more work in this area. I would urge you to take a look at some of the objections listed by the Open Rights Group before you decide [2]

Yours sincerely,

Ben Weiner

References:
[1]http://www.theyworkforyou.com/lords/?id=2009-11-25a.377.5&s=digital+economy+speaker%3A13699#g417.0

[2] http://www.openrightsgroup.org/blog/2009/write-to-the-lords-today

Lord Puttnam, former film maker, is deputy chairman of Channel 4.

Edit: reformat due to abysmal HTML from Blogger :-(

Web fonts podcast

2009-11-27T23:57:00.002Z

I respond loudly and with vigour and some inaccuracy to quiet questioning from Garrick Van Buren.

Understanding Web Fonts: book proposal

2009-10-30T08:34:00.005Z

Comments on this proposal appreciated.

Synopsis

This short book will offer a guide to anyone who wants to understand and implement web fonts. Its practical section will concentrate on WOFF fonts which seem overwhelmingly likely to offer the most future-aware and ideology-free format. The goal will be to give web designers and developers a clear understanding of the new technology, so they know when and how to implement it.

The Project

I would like to invite collaborators and contributors personally, but anyone who wants to contribute is welcome to contact me as project originator so we can compare notes.

The text will be written online, although see http://www.socialbysocial.com/book/creation-social-by-social

It will be published under GNU Free Documentation License, v1.3 or later, and the copyright will be held by ‘Contributors to the Webfonts book’.

A commercial publishing agreement will be sought. This would be comparable to the agreements for books like Version control with Subversion and Pro Git. If the book is published commercially then any royalties will be distributed between contributors involved in each static release (== print edition). http://www.gnu.org/copyleft/fdl.html

To set the topic in context it will give in about 10,000 words

an outline of the history of the technology
an explanation of advantages, both perceived and real
a guide to the history of typeface publishing, to explain where the special problems come from
a concise set of use cases for web fonts:
- graphic design and branding considerations (the major commercial drivers)
- academic research into ancient languages and transliterated documents that use obsolete characters
- minority languages (those poorly supported by operating systems: the Indian subcontinent, for example)

The main text of the book, which is at a first guess estimated at 15,000 words or so, would provide

technical info about the preferred format, WOFF, as a set of references to, and elaboration of, documentation emerging online (eg the Kew/Leming/van Blokland WOFF spec, the W3C Web Fonts group charter and Candidate Recommendation pages)
howtos for authors wanting to
- use cloud font services like Typekit
- make WOFF fonts, including advice on subsetting for lower bandwidth/faster rendering
- host web fonts
- create style rules that encompass the widest range of user agents
- understand their legal rights and obligations when using commercial web fonts

Then, as it is very early days, we may speculate briefly on what will happen in the future in about 5,000 words:

impact on the web
impact on the font publishing world
Fine typography in user agents, eg the Mozilla demo by Jonathan Kew. http://bit.ly/3sXGLh
an opportunity for free/libre fonts?

References

Discussion/analysis: oldish but set context

Edit: corrected heading hierarchy

Green lights ahead for WOFF web fonts

2009-10-21T09:32:00.006Z

OK, following an entertaining summer of debate on the www-fonts list at W3C and elsewhere, things are moving along nicely now in the world of web fonts. Yesterday saw two significant announcements:

the W3C Web Fonts Working Group gained a charter, albeit a little draft, which means a group should be getting together soon to formalise the Recommendation on the technical details of web fonts
over in the real world, Mozilla published a list of 33 proprietary font software vendors (‘typefoundries’) who endorse the WOFF format developed initially by Tal Leming and Erik van Blokland, later by Jonathan Kew and John Daggett.

A key benefit of WOFF is that it carries metadata and this metadata doesn’t have to stick to a rigid set of predefined fields. That leaves the door open for people to expose some of the information about publishing and rights which is usually hidden: stuff like who drew the outlines and when. In the case of remixed permissively-licensed fonts (using the SIL OFL for example) it could also include a log of which other open fonts were mixed, matched and modified to produce the end result. Think Zlicko, Deck, Barnbrook and co for the network age.

And yes, Microsoft is coming to the party. Hopefully without stinkbombs.

Royal dog in the manger

2009-10-05T13:46:00.005Z

I know that Royal Mail is a sort-of commercial entity (well, it’s at least partially a limited company), and I realise that having belatedly realised how useful postcodes are in an information-rich and hence information-dependent society they’re only doing what anyone would do when they act to prevent the use of the postcode database they inherited from more socially enlightened days without agreement of a substantial license fee. I also realise that its maintenance costs time, money and some expertise. I also realise that operating a system that circumvents the protection they’ve placed on the data will trigger cease-and-desist letters, however smartly done.

On the other hand, it is undoubtedly in the national interest for this data to be more widely available, especially when we consider the potential for services that might cater for the less-privileged rather than to big business budgets. I’d vote for a publicly-owned Royal Mail with publicly-accessible postcode-to-geocode data; that might be the fairest bargain to strike given that Royal Mail sells services based around the data it keeps about us all, without our conscious agreement.

Standing on the shoulders of web pages

2009-09-25T07:57:00.006Z

Just to record that something I have wanted to have for a while now exists in the form of a sidebar from Google. They call it ‘Sidewiki’ and it allows people to start a stream of comments alongside any web page. In the background this gets stored somewhere by Google along with all the other stuff they seem to hang onto. I can’t judge it on a screenshot or a thirty-second video but it looks as if the weakness, if there is one, is the separation of the two streams (original, critique) into different areas of the page. This is OK for general commenting but does not support remarks specific to a single element (a word or a sentence for example). I’d like to be able to add simulated post-it notes in place on the page, but of course that would rapidly obliterate the original content. Perhaps there should be lines to connect comments with source text? No idea how that could be done, but hey.

And then of course there is the potential for drive-by abuse. We shall see how that one works out.

Thoughts: what happens when people use the Sidewiki to add comments to a blog post which has already has a comments facility? Are people who don’t have the Sidewiki installed going to be at a significant disadvantage: will it become an essential browser component?

Thanks to Slashdot for the nod.

Truly an excellent medley

2009-09-23T23:38:00.004Z

I’ve just attended this concert [the one on 23 September that is] given by Passamezzo and merely wish to record how much I enjoyed it (very much) and urge others to see them. Seventeenth century music is currently rocking my little world.

Hope I’m not becoming a red Tory, but…

2009-09-16T13:37:00.006Z

I recently met the writer Adam Wishart and today found a link on his blog to an episode of Start the week from mid-June 2009 in which he participated. I enjoyed the programme enough to listen to it all, and hence it seems worthy of a post here.

It was explained that Adam’s recent documentary The price of life examined the way in which groups lobby the UK government agency NICE, which vets new drugs, for their cures of choice to be underwritten by the NHS. In the light of more recent news from the USA it is rather good to think that in the UK the argument is over the dispensing of a social service rather than whether such a service should be brought into being.

In general the conversation was around the issues of entitlement and liberty that are I suppose the characteristic fodder of the thinking middle class. But it contained a few thoughts that chimed with my own, particularly from Phillip Blond, a political philosopher who calls himself a ‘red Tory’. Amongst other things, he advocates supermarkets would be best broken up and their business returned to independent shop keepers. Further than I would go, but interesting. His Wikipedia profile provides more quotes.

Climate Wake-up Call on 21 September 2009

2009-09-16T07:01:00.002Z

A rather nice use of Google Maps in a good cause. I’m intending to attend the Reading meet-up which is scheduled for 1310 – true to form for Reading it is outside John Lewis in Broad Street.

http://www.avaaz.org/en/tcktcktck_map/

Hardware refresh goes on order

2009-09-15T08:02:00.004Z

In the end the decision was mostly about the quality and characteristics of the hardware, and in particular the mouse pad which I use for vector editing and page layout. It needs to be large and come to hand immediately. I also thought hard about the OS and I will be installing Ubuntu alongside (plus countless virtual machines, to be sure…). But even in its weird coding ghetto and even if my Adobe applications are DOA, I still have to acknowledge Apple is the only operating system provider, proprietary or otherwise, that can get me working as happily as it does. And the current machine has been staunchly reliable in the face of serious maltreatment. I thought for over a year about a new machine, then waited till money was tight, so I hope I made the right decision.

Number 10 responds to the petition to save Bletchley Park

2009-08-26T20:56:00.005Z

Bletchley Park near Milton Keynes, about 50 miles north of London, was the British nerve centre of second world war military code-breaking and the birthplace of the modern computer. Within the last twenty years its historical significance has been recognised and the remains of the wartime site now has a museum dedicated to explaining its past, as well as the reconstructed Colossus hosted by the National Museum of Computing (passim). But funds are tight and the supporting organisations failed to get grants to ensure the site’s survival in the medium or long term. As linked through from this page, a petition was lodged on 28 May with 10 Downing Street via their excellent MySociety–built online petition page. The official response was published today.

Although HMG itself is not going to intervene directly, Downing Street points out that things are being done by a variety of bodies such as English Heritage and the Museums, Libraries and Archives Council to help Bletchley Park survive. The actions so far include £900,000 of aid towards critical restoration work. Votes of confidence. I look forward to a brighter future for this strange and unique place.

The Number 10 response to the petition

SVG for IE

2009-08-24T07:14:00.007Z

As (slightly innaccurately?) posted on Slashdot, a Google Code project to create an SVG plugin for Internet Explorer has gone public with a little video showing what their software can do.

Actually, as SVG is vanishingly rare on the web I think we could all do with a refresher. If you watch the video turn the sound off though.

PS there’s a natty embedded SVG and Javascript logo animation on my homepage. The SVG is loaded into the DOM using some rather flaky AJAX. Been there about three years and nobody’s ever asked me how it works ;-)

Edit: oops, incorrect syntax in that first link tag

Another edit: should have credited Slashdot for the link.

Computers find a new home

2009-08-03T09:18:00.006Z

Last weekend I foisted a couple of my old computers (old as in fifteen years, not fifty years) onto the National Museum of Computing at Bletchley Park. It was great to meet some of the volunteers including Pixelh8, who seems to be capable of making any old computer, no matter how brusque its exterior, sing sweetly.

NB. Don’t take stuff to Bletchley Park unless you have contacted the Museum to check that they can accept it. There is precious little space available for storage and, as ever in the museums and archives sector, little money to pay for it.

Type-loving US lawyers will need big monitors

2009-07-28T08:05:00.004Z

OK, that’s a little beside the point. But you will need plenty of pixels to read through this new and commendable introduction to typography that has been provided by an lawyer, ex-typographer, to other lawyers not ex-typographer from the kindness of his heart.

NB I’m a real enthusiast of web sites with large type. Let’s have more!

Link via the LETPRESS list.

Sparks to fly

2009-07-23T07:02:00.003Z

Good to hear that major UK rail electrification plans are to go ahead. This is a significant government policy change. Lots of metalwork will be required. Is this work for the Teeside steelworks?

And I wonder what will happen to the rest of the related proposals in Network Rail’s Rail Utilisation Strategy document?

Speaking of TTF web fonts …

2009-07-17T14:14:00.002Z

Here’s a commercial font-hosting service that is just starting up

Updating Ubuntu major version in a Virtual Box VM

2009-07-16T08:37:00.004Z

Yesterday I upgraded from Ubuntu 8.10 to 9.04 using the system’s Update Manager (System>Administration>Update Manager). This worked perfectly, which is impressive considering I upgraded to a new major version of the software, but because the Ubuntu in question is a VirtualBox VM the new OS version’s X11 video software wouldn’t work properly. It complained that it couldn’t find a suitable driver and it would not work in low resolution mode.

The fix was to log in to the console (a dialog box reporting the error offers you this option), then mount the VirtualBox kernel additions ISO image (which is called VBoxGuestAdditions.iso and is, I think, something internal to VirtualBox and hence always associated with the VM by default). The ISO image contains scripts called VBoxLinuxAdditions-<arch>.run [the one suitable for my hardware is VBoxLinuxAdditions-x86.run], which you can run to install the twiddly bits, hacks, patches, extensions or whatever that enable video and probably a lot more too.


$ mount /media/cdrom
$ cd /media/cdrom
$ sudo ./VBoxLinuxAdditions-x86.run
[script runs]
$ sudo shutdown -r now

Good news for web designers: acceptable web-linked fonts are here

2009-07-15T08:29:00.005Z

A rumbling debate on www-font, the W3C mailing list forum for discussion of how linked, downloadable fonts should be implemented, has been cut short by the action of commercial font publishers. At least one significant player is now issuing TTF fonts with web server licences, which is effectively an end-run around the discussion because it means that the dual-format (TTF/OTF and EOT) question now has an answer from the font software publishing world as well as implementations from the browser authors and publishers. These two groups held the power between them, with web designers and developers yapping and clamouring to more or less effect in the middle. Now the matter is resolved you can expect to be able to license TTF/OTF and EOT versions of many high-quality commercial fonts very, very soon. I’m not sure whether anybody lost out here, but I think commercial font publishers will fall over themselves to avoid being the last ones to make their libraries available.

Personally I am disappointed because I was hoping that the debate might move forward the concept of licence expression, where information about creator and publisher of linked items in a page (like video, photography, typefaces, etc) is available to the user through a machine-comprehendable format that is partially or completely wrapped around the media themselves. It would bring to the surface a load of people working in the second tier of the web: the designer and publisher of a typeface, the photographer who took the spectacular travel picture, the painter of a portrait and its licensor (typically a gallery, museum or private owner), the writer and producer of a video, etc as well as the people who design and code web sites. Such a system would play well to the idea of a semantic web. Tom Lord’s proposal, MAME, now seems likely to stay that way because, as clearly demonstrated in the font licensing discussion, nobody cares.

How it works

Here is a method that might be used to prevent TTF/OTF font files being ‘stolen’ (used outside the terms of the licence under which they have been provided) and hence could be acceptable to a commercial font software publisher:

Server implements referer checking and denies the requested download if the check fails (making it hard for people to simply look for the URLs of the font files and download them by typing the URL into a browser)
Font name table in the linked font file is made unintelligible by the publisher (the name table is critical for the font to be added to UI menus, caches etc on the operating system but, because all the info is replicated in the @font-face rules, browsers don’t need it)

Nowhere in any specification is the referer check or the obfuscation of the font file made mandatory; instead they are conditions of doing business with the font software publishers and you can always use permissively-licensed fonts instead if they meet your requirements.