The English Cemetery

Deeply moved by the story and courage of Ivan Noble. And a striking coincidence: my friend Joxe Aranzabal, also a journalist, began writing in Internet about his newlyfound cancer just two days after Ivan, in september 2002.

Joxe has struggled with the cancer, and has prevailed. He has left a message at Ivan's last post, don't know if the BBC will let it public or not. However, here's what Joxe has written:

On September 11th, 2002, I was found to have a Pholicular Linphoma, a kind of cancer that appears in the blood. A couple of days later, I started writing what was to become a series of articles in Sustatu.com, a collective blog with information in Basque language, where I narrated in 17 chapters all the circumstances surrounding my illness. It took a couple of years to get back to work, at Mondragon University. The 22nd of this month I will publish a book (The Doctor Told Me I Had Cancer), a compilation of all the articles written for Sustatu. When we present the book, I will have you in mind. Be strong.

Besides his writings at Sustatu.com Joxe has a personal blog now, Faroa, the Lighthouse in Basque.

First there was Planet Planet, Python package to aggregate feeds. Now there's Planet Express (just version 0.1 so far), to aggregate feeds and re-load them into a given blog.

In this case, the author, a well known Spanish geek, RVR, has made it so it can feed automatically Typepad blogs. So, you have a metablog there: take feeds from distinct sources, and re-package them as a new site. Working example, Planet Spanglish.

I wonder if it could be adapted to feed a Coreblog, instead of just Typepad blogs... After all, it's made of the same raw material: Python

We have hacked the Squishdot based Basque weblog Sustatu so it now feeds Technorati and its categories appear at blog tags in the tag search application unveiled yesterday.

Results already are visibe, as with this two obvious tags:

• http://www.technorati.com/tag/musika
• http://www.technorati.com/tag/kultura

Other Basque blogs made with WordPress show up, as well as postings in languages in which musika and kultura mean what you know.

As for The English Cemetery, it's too modified, I'm afraid, for the Coreblog tag hack that we deployed yesterday. I will try with manually entered Technorati tags, as they advise in their website.

So, let's try with coreblog basque zope

Technorati has released today a very interesting service. They've developed a tag based search application based on Flickr tags, del.icio.us tags and blog post's categories.

Two examples with tags google or basque

More info about Technorati tags

As for Coreblog, a little hack by Erral, that I post here with his permission:

COREBlog doesn't give info about our posts' categories in RDF files, so we need to change some lines of code in 2 files to show our posts in that service.

Let's go to change them ;)

Open rdf10_xml via ZMI:

Find this line:

  ...  <dc:subject />
 ... 

Delete it and put these lines:

   ...   <dtml-in entry_category_list>     <dc:subject><dtml-var name></dc:subject>   </dtml-in>
  ... 

Save it!

Now, the second step: open rdf91_xml via ZMI:

Find this line (it's the 3rd one):

  ...  <rss version="0.91">  ... 

And change it:

  ...  <rss version="0.91" xmlns:dc="http://purl.org/dc/elements/1.1/">  ... 

Then find these lines:

  <dtml-in "rev_day_entry_items(count=top_days)">  <item>
   <title><dtml-var title></title>    <link><dtml-var blogurl>/<dtml-var id></link>   </item>   </dtml-in>
  

And change them:

  <dtml-in "rev_day_entry_items(count=top_days)">  <item>   <title><dtml-var title></title>   <link><dtml-var blogurl>/<dtml-var id></link>
  <dtml-in entry_category_list>     <dc:subject><dtml-var name></dc:subject>    </dtml-in>   </item>   </dtml-in>
  

Save it!

Now, we've to configure our COREBlog to ping Technorati each time we publish a post. Go to "Settings" tab, and add http://rpc.technorati.com/rpc/ping at "PING Servers".

This actually works. Erral's blog is in Basque and Spanish and his spanish category/tags already show at Technorati as with ajedrez and varios

Brilliant idea: Taggle, an hypothetical search engine of the future, to search into folksonomies.

But Taggle will not work with a traditional webcrawler or robot behind, that's for sure. Del.icio.us has a strict robots.txt file forbidding access to robots...

However I do believe that Taggle will exist, based in web services, the APIs of providers, feed aggregation or other non-crawler method.

The bug I commented the other day, has re-surfaced with the new year. If there's a year "gap" in the archives, a year with no messages in it, the archives module of Coreblog shown at the sidebox will not work. It just happens that if it's a new year, and we haven't posted any entry to our coreblogs, then the bug works, and we are left with no archives apparently.

Look at Trond's "blog."http://www.bek.no/Members/lossius/lostblog He's been early this year, and his sidebox does hove archives. But Tom Lazar's been lazy this year so far (as of this morning) so his blog shows an empty archive...

Provisional solution: post a message every 1st of January.

But, obviously, this is a bug to solve in the product.

Happy new year to all!

Robin, a bilingual blogger, has invented DURL. You enter a url to retrieve information about people who delicious'ed it. For instance, tools.waglo.com/durl/http://www.ysearchblog.com/ and the result page has its own RSS, so you can track who bookmarks what at del.icio.us

Excellent idea!

One addition that I think could expand the capabilities of DURL or of the URL features of del.icio.us would be this: URL listings and RSS feeds by partial URLs.

I mean: Now DURL or del.icio.us offer information about a precise URL:

 http://del.icio.us/url?url=http://www.ysearchblog.com/
 http://tools.waglo.com/durl/http://www.ysearchblog.com/

But I would like an extended feature like:

 /url?url_extended=http://www.ysearchblog.com/
 or /durl_extended/http://www.ysearchblog.com/

That would list delicious'ed references to all URL's include that include that domain, for instance if anyone referenced to this or to this one

Such a tool would certainly enhance the ego of bloggers like me, that would immediatly subscribe to a feed from /durl_extended/http://www.eibar.org/blogak/luistxo/ to see who bookmarks which entry (and every entry) of my blog...

Of course, durl_extended should not substitute the actual DURL: it would be a different RSS feed the one provided by:

 /durl/http://www.ysearchblog.com/ 
 (bookmarks referring to that URL alone)

and the one by:

 /durl_extended/http://www.ysearchblog.com/ 
 (bookmarks referring to all subdirectories or else)

This is a demo of a OAI harvester and data-provider. Its purpose is showing the capabilities of such a system, in this case offering a multilingual interface, applied to the context of the Basque Country and the Basque-language cultural production.

We used free software to do it, mainly ZOpenArchives and Localizer (Zope products) to produce the multilingual interface. ZOpenArchives is an OAI implementation for Zope that transforms Zope into an OAI Data Provider and Aggregator. Collaborative project

This Demo is the result of a collaborative project between CodeSyntax and DELi, a research group from the University of Deusto. The demo shows the capacity of OAI to function both as a data harvester (that includes a search interface) and as a data provider

The collection of OAI records used for this demo were kindly provided by several Basque institutions and publishers. Beware that most of the references correspond to Basque records, and general-purpose searches in English will not produce too much results... Follow the tips at the search page to obtain meaningful results.

The Basque context

Our purpose is to show in our local context that a Digital Library based on OAI's makes much sense. A minority language, Basque, where resources are many times scarce: it makes sense to promote a distributed model for a digital library. In our vision, a Basque digital library should not be like this:

Not a central mainframe center, with gigantic centralised resources that everyone else should fill following N protocols... Rather, that Library should follow this model:

That's the hydrography map of a Basque region. Each council or county has its own responsability to maintain its own realtively small reservoirs. Then, there's a network connecting them. No need to feed a central system: anytime it rains here or there, water pours on the general system.

That could be the model for a Library, yes! Each one publishes its own records according to its resources, following the standard, and the network becomes the Library then.

What RSS is to news, OAI can be for books, cultural items, documentation at large... I am convinced of that. More about OAI at OpenArchives.org

Thanks

The following Basque organisations have contributed data to this demo:

• Badihardugu euskara elkartea, an association devoted to the study of local varieties of the Basque language.
• Gaztelupeko Hotsak, a music recording company.
• Gerediaga elkartea organiser of the Durango Fair.
• Inguma, a scientific database from the Basque Summer University Programme UEU.
• Megadenda.com, an online vendor set up by the book-distributing company Zabaltzen. The Basque publisher Elkar and its related software company Plazagunea have also contributed.
• Ibinagabeitia proiektua, a Basque online archive of literary journals developed by the publishing house Susa together with the ASP software company.
• Lanbide Ekimena, which is a documentation project for vocational training in the Basque Country, and is supported by several institutions: Hetel, Ikaslan, the Basque Government, BBK, Elhuyar and the regional administrations of Gipuzkoa and Bizkaia.

This OAI demo and the OAC project have the suppor of the Saiotek Research Programme from the Industry Department of the Basque Government.

It's a nice feature in Coreblog that one you add an entry through the ZMI, you can change the date of the entry, putting your own "Entry date&time" in the field for it.

That way, you may post a message scheduled for next sunday, and it will be on the front page that day, not before. You may also rescue all writings from the past, and build an online archive of yours with the Coreblog.

However, there's a problem:

• Once an entry is added, no way to change its date format. Suppose you made a mistake with "Entry date&time": you can't edit that thereafter. Another time you need to do that is that, for some reason, a message you wrote one month ago becomes pertinent again, and you want it to be again at the frontpage with and updated date. No way.

You may change the text, title, edit comments, encodings, but nothing with the date. Only way is re-copying the content, deleting the entry, and add a new entry.

So, my question is, perhaps someone knows a trick at edit dates?

Year-by-year archive

On the other hand, I discovered a curious thing with past messages:

1. You have entries of 2004 at your blog, the months that you have posts appear at the Archives section at the sidebar.
2. But now you add past messages from may 2002 and october 2002: those months will not show.
3. Another try: now you add a past message with date for june 2003: now the bug's corrected, not only June 2003, but October 2002 and May 2002 will appear also.

So, year-by-year continuity seems to be needed so the Archives section may work. If there's a discontinuity (one year with no posts), entries of the previous years will not appear at the Archive monthlist.

I have just discovered Blogplot

It seems to be around for a year os so... But I didn't notice it.

Blogplot.com is now hosting free CoreBlog accounts. You will have access to all the standard features of CoreBlog with the capability to customize to your needs. You will also have access to all Zope products on our server in your CoreBlog instance. To get started simply email us with a subject line "Free CoreBlog Account" along with domain name information. You may register your own domain name and we will gladly manage your dns, or you can simply use a sub-domain of Blogplot.com (i.e. yourname.blogplot.com).

There are already a bunch of Coreblogs hosted there

well, I think it's worth to know there's Coreblog world outside Japan...

It's the second Coreblog community that I know, after our own at Eibar.org but well, theirs is in English and have more independent URL's than ours. However, our Coreblogs have a small fixed content editor for extra pages that comes handy.