Reclaim Your Social Media Content

From Reclaim Social: “Depending on your personal level of addiction, everyday you create, curate and share things on the internet. Great. You use about 384 social networks, that just don’t belong to you. Still great. Sort of, who wants to own facebook anyway. But if you search anything you shared or liked a year ago, you’re lost. If your account is suspended, your data may be lost completely. Reclaim Social is a wordpress based concept, consisting of some plugins and scripts. It allows you to mirror and store your content and activities around the web on your own blog“.

The Reclaim software is far from finished, so if you know anything about WordPress plugin development you can lend a hand to this open source project started by Sascha Lobo and Felix Schwenzel.

I like the concept, even though I don’t need it for myself: I have almost all of what I ‘produced’ right here on my blog (except for afew hundred not so interesting tweets from more than a year ago). But being able to keep a copy of what you publish is a good idea, in my view. Dave Winer proposes to author your stuff in a single tool under your control, then publish wherever you fancy. Reclaim proposes the inverse way: publish where you want, then copy it all into your blog. Conceptually, Dave’s approach has the advantage; in practice, many people may well find a tool like Reclaim Social simpler to use. And WordPress is a good platform for such a tool, if the reclaimed content can be a part of the normal backup/restore tool in WP.

Read Full Post »

How Hard Is Content Management Anyway?

A few days ago, Daniel Greenfeld wrote a blog post on the “Tools we used to write Two Scoops of Django“. Mentioning the words “tools” and “Django” in a single phrase is sure to get my attention, and the post did not disappoint. What fascinated me most, however, was not the “tools” part nor the “Django” part. The story is a good example to explain which difficulties show up in the process of creating and managing good content – even when it is “just” the writing of a book and its publication in multiple formats.

If you read the story, you will notice that

• the authors started using familiar tools, like Python’s reStructuredText and Google Documents;
• they then noticed that those tools did not measure up to their requirements, and had to switch again and again – until they ended up with LaTeX, allowing them to clearly structure the content and only then apply the formatting and layout for a paper version;
• e-book formats are a world on their own, and it’s a complex world with many pitfalls;
• technical content is never finished, so it requires forethought and planning to keep up with reality during the lifetime of your document.

All this about a relatively simple thing as a book, and by that I mean no disrespect to the authors. But assembling and maintaining say a website with hundreds of pages will be even harder, since such an endeavour add factors such as user management, target audiences, multiple authors, possibly multiple languages, categorization of pages, an editorial policy, and more – and that’s not mentioning a series of technical aspects. Enterprise content management ups the ante again, sometimes by several factors. So, yes, Content Management can be hard, very hard, to do right.

* * *

And what about the book, you ask? Well, if I’m ever going to do anything serious with Django I’ll need a copy of “Two Scoops of Django“. The sample chapter was a bit over my head, since I’m an absolute Django beginner, but the patterns and tips seemed logical and well explained. The writing is clear and sensible. The only thing I did not like is the fact that code samples are sometimes being split over two pages. That should not happen except when the code is longer than what fits on a single page, and even then I would prefer the code to be rewritten and packaged in smaller blocks. But all in all, this looks like a worthwhile book.

Read Full Post »

An Unexpected Appearance Of AI And Governance

Wired’s ‘The New Spoiler Culture: Game of Thrones and the Fight to Live Uninformed‘ may not be what you want to read if you’re a GoT fan. But if you work in the content management business, you should pay attention and note that the fan websites mentioned implement a finely tuned information architecture in addition to governance policies! How else do you describe their efforts to make sure that their intended public sees exactly that part of the site that it wants to see?

Working out a specific information architecture and writing down the corresponding governance rules isn’t limited to the content management that goes on – or should go on – within the enterprise. Both activities (as well as many others) must be taken in to account for any serious content management project.

Read Full Post »

Tips To Strengthen The Security of Your WordPress Site

Recently hackers have launched large-scale attacks on WordPress sites. If you run your own WordPress instance you could do worse than read SitePoint’s “WordPress Security“. Don’t just sit there, read it, and act!

Read Full Post »

Wiki’s Are Still A Relevant Content Management System Variant

Wiki’s are no longer the spearpoint of the ‘social media’ that they were a few years ago. That does not mean they have gone away or, worse, have become irrelevant. It’s good to see that a major player like Ars Technica publishes a nice Mediawiki installation manual: “Web Served 7: Wiki wiki wiki!“.

The rest of the ‘Web served’ series – on how to setup and use a secure webserver – is worth reading as well, if you have little or no experience with the subject.

Read Full Post »

13 Years And Still Blogging

Today, this blog has been running for thirteen years. No, I wasn’t the first blogger, not even in Belgium. Nevertheless, thirteen years merit a little celebration, especially now that the content of my original blog has completely been replicated on this site! That has taken a long time, but as I explained earlier: I did it by hand, re-reading my writings and trying to make sure that the links and pictures were still what they are supposed to be. That wasn’t always the case, so you’ll find some links replaced with links to the Internet Archive (the Wayback Machine), some crossed-out, and some in posts put in category ‘Linkrot‘ (and I probably missed a few broken links – don’t hesitate to point them out if you encounter them). All in all, I didn’t lose too much, except for the research time required to try and replace lost target pages. There’s a lesson to be learned from that work: when I show images, I will copy a 400px wide “thumbnail” from the original and link from that thumbnail to the original. So if the original gets lost in the depths of the Web, I’ll still have something to show. Similarly for texts I link to: when discussing a particular point of view, I’ll copy the essential parts as a quote and link to the original rather than just add the link .

A lot happened in those thirteen years. Don’t read too much in them, but I do want to share the numbers with you. The number of blog posts on this blog, per year, to be exact. Can you imagine that I once, in 2004, found the time to blog almost every day (if you don’t count the weekends)? Luckily for all of us, I decided that quality beats quantity ;-)

Year	Posts
1999	2
2000	54
2001	94
2002	161
2003	231
2004	282
2005	208
2006	178
2007	136
2008	117
2009	115
2010	128
2011	141
2012 (up to and including November)	150

One thing is certain: the future will bring more posts ;-)

Read Full Post »

Three Years Of WordPess

Three years of WordPress, that is what today means for this blog. WordPress has shown itself splendidly up to the task of managing my blog, and in those three years I have learned that WP is more than a simple blogging tool: it’s a serious CMS, worthy of consideration for anyone who wants to build and run a website. I’m writing this while riding the train to work, and that proves that WP has included the mobile revolution in its solution package – nice.

I don’t follow the WP blog, so I don’t know what the WP developers are up to. But I am curious to see how WP will evolve in the coming years!

Read Full Post »

My Blog Migration Is Finished – At Last!

It took me nearly three years (elapsed time, not working days) to migrate the content of my blog from nukleos.editthispage.com to this platform. I’m glad it’s done, especially since it took so long. Why did it take so long?

Let me explain: three years ago, I thought briefly about automating the migration. The Manila platform of editthispage.com can export a complete site, and WordPress is perfectly capable of importing a site packaged in an XML file. But this route would have required me to relearn Frontier/Manila skills which I had not used for many years, after using Frontier for one of the first websites within the company of my employer in the early nineties. Additionally, automating the migration would not have helped me in discovering typos and failed links.

So I started the migration by hand, copying the text of each blog post , then pasting it into a new WordPress post, resetting the date, checking links and possibly replacing them with relocated URLs or links to the Wayback Machine, adding categories, inserting images, etc. All in all, if my counting is correct, that made for 1561 posts to be copied – and let me assure you that there was no way I could do that for hours at a time: it’s just too tedious and stressing at the same time. Thus the pace of migration was at time fast, and sometimes I didn’t get around to doing it for weeks or months.

But as of now, almost to the day three years after I started the migration, all of my blog posts from the past 13 years are reunited in a single, performant tool. At the same time, those three years made me rediscover not just a lot of links I had forgotten, but also a high number of my own writings, also hidden in far corners of my mind. Maybe it’s because I’m a historian by education, but I liked that journey into the past. All that stuff is now back together in a single searchable form, ready for instant rediscovery – I’m sure that will pay off one of these days.

Read Full Post »

W3C Wiki Documents HTML5, CSS3, And More

The W3C, the standards body that oversees the development of HTML, is starting a Wiki to (hopefully) create a definitive, up-to-date collection of documentation on all the web standards (and more). In the words of Webmonkey: “The W3C has managed to bring together some of the biggest names on the web to help create Web Platform Docs. Representatives from Opera, Adobe, Facebook, Google, Microsoft, Mozilla and Nokia will all be lending their expertise to the new site“.

Of course, the really interesting part – for me, that is – is the Wiki that will encompass all those Web Platform Docs. I do wonder how much “non-expert” material will be added to the site – writing documentation is not the same as writing code, as evidenced by the lack of decent documentation in many software projects. Then again, not just developers can (should) add their expertise.

I guess the W3C will be exercising some kind of editorial supervision. To their credit, the creators of the site have clearly indicated how anyone can help, even if you’re unfamiliar with the technologies mentioned – that’s a good idea for any Wiki, by the way.

Read Full Post »

JCR And CMIS: Complementary Technologies

I attended an SAI seminar on CMIS yesterday, starring a presentation by Joeri Samson and Stijn Van Vreckem from XAOP. The term ‘CMIS’ (“Content Management Interoperability Services“) is gaining notoriety these days, now that Enterprise Content Management (ECM) systems are being contemplated and installed in an increasing number of enterprises and public services. CMIS can be an effective tool to bridge old and new ECMs, as well as a means to englobe multiple compliant ECMs for diverse purposes like search, archiving, etc.

One of the questions that popped up after the presentation was: How should we compare CMIS to the Java Content Repository (JCR, specified in JSR 170 and JSR 283)? I knew that the JCR is CMIS-compatible, but just to be sure I did some checking today. And the answer to the question is, of course, that the two are not competitors, but complementary technologies. In short:

• The JCR is, as the name says, a generic ‘content repository’ (a data store) with an extensive Java API.
• CMIS is essentially a protocol to talk to any CMIS-compliant content repository, not just in Java, but in any of many programming languages.

And yes, any good JCR implementation is CMIS-compliant: here’s David Nuescheler introducing both technologies. No, you won’t be able to do all that is possible in the JCR through a CMIS browser, but you’ll be able to consult and update objects in the repository, just like for any other CMIS-compliant repository. CMIS is still evolving, by the way: version 2 of the standard is being validated these days, so I suppose an more powerful CMIS is coming in the very near future.

Read Full Post »

Word Of The Week: Mapocalypse

Of course, Robert Cringely writes about the snafu with Apple’s iOS6 maps application troubles: “Apple’s Maps is so spectacularly bad it’s inspired its own Internet meme. Funny? You bet“.

But as always, it pays to check your facts before publishing. Even Google Maps knows about the “tiny hamlet in Austria where every street name starts with the F word” – because that hamlet does exist in the real world! Just check the Wikipedia: Fucking, Austria.

Then ask yourself: why does Google Maps call the streets in that hamlet “Hucking” ? Creative spelling indeed!

Read Full Post »

Foundations For A Wiki

WikiNotes is a wiki-based note-sharing platform created to facilitate student collaboration (beware: the site can be slow!). Basically it’s a system that allows students (and others ;-) to publish course notes, questions from examinations, study notes, etc. The current version of WikiNotes is based on the classic MediaWiki-software that also powers the WikiPedia. But a small team of student developers is rewriting the system from scratch. A beta version can be explored at http://beta.wikinotes.ca/ – and its source code is published under the GPLv3 on GitHub.

Why do I mention this project? First of all: I am a Wiki fan ;-)

Secondly, it’s a good example of what a Wiki can be used for: the creation of a public body of knowledge that is fed and maintained by anyone who wants to help. Of course, not all courses are equally well documented, but since a Wiki can be edited anytime that might change rapidly (or never ;-). Not everyone was (is) convinced of the value of such an initiative – users of WikiNotes should be aware of possible copyright issues, to name the most frequently mentioned counter-argument –  but I have found several pleas for a more nuanced approach that can benefit both the students and the University. Check out this editorial from the McGill Tribune for suggestions on how to govern the WikiNotes site.

Thirdly, the new version is being written in Python and created on top of the Django framework. I find it stimulating to see the progress in the development of this application, from a developers prespective.

Last but not least, this Wiki started out as a student initiative – there are clearly smart students at McGill!

At the same time, this project illustrates an essential aspect of any content management project. Before choosing any CMS, or writing your own, you have to analyse your requirements. You have to know what you will be dealing with: what type of “content” (text, image, video, …) about which subjects will you be handling; what metadata do you need for each content type; who is going to handle that content, and what will they be doing exactly; who is the target audience for the content, and what are their expectations; etc. Without clear answers to these and more questions you will never be able to set up a suitable CMS, let alone claim a succesful implementation.

PS. This is why I fear that the brand new Brussels Wiki (‘Wikibru, the wiki of the City of Brussels‘) will be a failure: a wiki needs more justification than the possibility “for people to add something to the site”…

Read Full Post »

Don’t Forget Your Past, Dear Website

As websurfers, we do not really notice the tools and technologies that drive the sites and applications we visit and use. Yet, the last decade has seen an enormous evolution: most current content management tools are relative newcomers, and to complicate matters most websites have frequently changed their structure, content and layout, be it within the same tool set or not.

I am still working on the last part of the migration of this blog from the “old” EditThisPage.com service to WordPress. I’m doing this manually, not just because I never took the time to convert the Manila backup into something WordPress can import, but also because it allows me correct spelling and other errors in the old version. And from time to time I check for linkrot too…

Today, I am migrating blog posts from March 2006, and the post from the 26th puzzled me (see the new version here). Clearly, the sentence “We love democracy, providing the Muslim nations elect the people we want” was wrong; the question was, did I miss-spell “provided” as “providing”, or did my source? So I clicked the link to the source, and got a 404 error message telling me the link in my post is dead.

The good news, of course, is that the website The Truth Seeker is still around. Even better: on their home page I spotted a link to their “Old site”. That old site does not look pretty, because there are obviously images missing. But it is still there, and changing my original link from “http://www.thetruthseeker.co.uk/article.asp?ID=4286” to “http://www.thetruthseeker.co.uk/oldsite/article.asp?ID=4286” brought me back to the source of my quote!

My conclusions from this little story:

1. Webmasters, please do not forget the older versions of your sites – just deleting them may create a lot of linkrot.
2. The Truth Seeker did the right thing and kept the older site;  too bad their missing page mechanism does not take the old links into account.
3. The spelling mistake I noted did exist in the source, so I’m leaving my quote as it was (but  I have added a small editorial indication explaining that I did not make the mistake).

Read Full Post »

Twitter Loses Again (Or Not?)

Until recently, I “shared” my tweets on my LinkedIn profile page. But I’m having more and more doubts about the value of Twitter in general – or perhaps I should say: in the “business model” that Twitter is currently rolling out. And I’m not alone in that respect: Dave Winer has been saying so since long, and more and more people have their own reason to think aloud about alternatives – like Brent Simmons.

Am I tempted to use LinkedIn as my Twitter client, as LinkedIn proposes? Not really. As I have said before: Twitter is not about conversations; there are far better tools than Twitter to converse. Besides, I’m not even sure the accounts I follow on Twitter are LinkedIn members…

Conclusion: I have removed my Twitter account from my LinkedIn profile page. I’m sure you’ll find other ways to start a conversation with me if and when the need arises ;-)

Read Full Post »

Is A Blogging Tool A Content Management Tool?

Of course blogging is part of the CMS space.

Dave Winer: “Manila was an early blogging platform, but it started out as a full newsroom server app, with discussion software, editorial roles, tons of features that you don’t see in blogging software. In a sense, blogging as an activity developed as Manila developed. It was launched in late 1999, and kept developing through early 2002.“

I “joined” the Manila universe late in 1999, through the EditThisPage.com site: you could start a site there for free (later on I paid for that service). Manila was indeed more than a simple blog. In fact, at the beginning it was rather hard to figure out how to blog with Manila: just look at the title of almost all of my pre-2008 blog entries. Yes, they all start with then name of the site, because to get the latest blog post on the home screen you had to “flip the homepage“, and since I wanted the title of my page to include the name of the site I had to add it to the story title…

Things have changed a lot in the CMS universe, but I’m glad that Userland (hi Dave!) offered me a fantastic introduction into content management so many years ago.

Read Full Post »