MJR's slef-reflections

Free software finally gets significant coverage on BBC TV's Click show this week, but I think it's very much Linux rather than GNU/Linux and free cost rather than freedom. They mentioned free security software and even raised the possibility of trojans, but didn't mention how free (as in freedom) software allows any random end-user to check or have it checked.

Quite a missed opportunity! However, Click has a regular letters section, so watch it (times below), email click@bbc.co.uk and see if we can get the free software view across.

The letters section this week seemed to be flaming proprietary SaaS social network site facebook for their pathetic default-permit approach to security of user details. I really think there's a role for something like noserub in free software social networking.

Click-UK is shown on BBC News Channel Saturday 1130, Sunday 0430 and 1130, Monday 0030 and Sunday 0430 on BBC-1 (times BST)

Click-World is shown Thursday 19:30 GMT, Repeated Friday 09:30 and 12:30 (Asia Pacific only), Saturdays 06:30, Mondays 15:30, Tuesdays 01:30 (not Asia Pacific, Middle East or South Asia) and 07:30 GMT

Anyone else see this?

"Here are three examples of rules that I think it's time to abandon. These particular examples are all about email.

1/ Top Posting [...]

2/ HTML Email [...]

3/ Reply-To On Mailing Lists [...]

So, yes, the barbarians are at the gate. The lunatics have taken over the asylum. Good ideas have been crushed by the number of people who don't understand them. But there's no point in complaining about it. You just have to accept it and move on."

-- Pointless Battles For Geeks, By Dave Cross

Unsurprisingly, given the above links to my site, I disagree with Dave Cross's conclusion, but I do agree with two aspects: battling is generally pointless and using hard rules about these things is unnecessary.

I have those pages on my website so that I can point to them when their broken emails aren't handled as expected. I use some aspects of them as scoring inputs in my mail filters. I don't use them as rules and I try not to complain about them too often.

Nevertheless, I still believe sending properly-trimmed plain text emails from a list-friendly email client is clearly best practice, to be recommended when someone asks why their email bad habits are causing them problems.

I'm convinced by The Argument for Posting Ahead [Network Blogging] so why didn't I set it up for my recent trip?

Well, a while ago, I had a bit of an email discussion with Steve about delayed posting features for Chronicle which I think ended with the addition of chronicle-spooler to the released files. I initially suggested ignoring a post when its Date is in the future, like Wordpress does, but I think that wouldn't be backwards-compatible and isn't as flexibile as it could be.

chronicle-spooler works by moving files from a spool dir into a live dir. I'd forgotten how much it confuses me when files on a "static" site start moving themselves around. It means I have to merge before upload, which isn't a big change, but still annoying.

So, I've added the following lines to chronicle to skip entries with a Publish header in the future:-

--- chronicle-2.7/bin/chronicle.orig	2008-05-15 10:13:55.000000000 +0100
+++ chronicle-2.7/bin/chronicle	2008-05-15 10:13:43.000000000 +0100
@@ -607,7 +607,8 @@
         #  Read the entry and store all the data away as a
         # hash element keyed upon the (unique) filename.
         #
-        $results{ $file } = readBlogEntry($file);
+        my $result =  readBlogEntry($file);
+        if ($result) { $results{ $file } = $result; }
     }
 
     #
@@ -1479,7 +1480,7 @@
     my $tags    = "";    # entry tags.
     my $body    = "";    # entry body.
     my $date    = "";    # entry date
-    my $publish = "";    # entry publish date - *ignored*

+    my $publish = "";    # entry publish date
 
     open( ENTRY, "<", $filename ) or die "Failed to read $filename $!";
     while ( my $line = <ENTRY> )
@@ -1527,6 +1528,14 @@
     }
     close(ENTRY);
 
+    # MJR - embargo stuff until its publish date.
+    # Steve recommends using chronicle-spooler,
+    # but I want uploaded files to stay where I put them,
+    # else I get my local copy confused.
+    if (($publish ne "") && (str2time($publish) > time())) {
+    	return 0;
+    }
+
     #
     #  Determine the input format to use.
     #

I think that's compatible with chronicle-spooler, too...

Anyone seen this before?

Today, I went to an event about the Sub-National Review Consultation (as a substitute for someone else AIUI).

I'd not heard about this before, but if you're in England and you've any interest in our regional planning system (which I think you should, if you have your main home here or run a business here), you have six weeks left to comment on the UK Government's suggested changes.

As I understand it, it will move the second-highest tier of planning control from democratically-accountable regional bodies to the business-led Regional Development Agencies, with some oversight by MPs and the very- indirectly-accountable council leaders. I've posted more detail on Co-opNet.

When I asked about local involvement and cooperatives, I was directed towards Local Strategic Partnerships, but I'm pessimistic about how easy it will be to influence regional planning through those: a few weeks ago, I was at the launch of the North Somerset Partnership Sustainable Community Strategy for 2008-2026.

It's a 72-page A4 glossy book which I've still not found time to read properly. I think the size says something about its sustainability. I've posted a little more detail on WsMForum.

I'll try to answer questions about either of them on this blog or those forums...

I'm just back from a trip to Cuba. Sorry for the abrupt disconnect. I expected to have limited connectivity there, whereas I actually had none at all except for 23 minutes! I hope the other members of the cooperative and the editorial team stepped in suitably well. I'll check in with them now and then start on the emails, but I wanted to put this broadcast out first for those who are watching closely and wondering...

I'm still alive, so it's past time to update the signature on my gpg key into next year. The key phrase in the handbook is:

"The expiration time is updated by deleting the old self-signature and adding a new self-signature."

but somehow I always have to look it up, so I thought I'd make a note of it here.

It looks like debian-keyring should update now I did --send-key to it, but I guess I'll find out in a few weeks.

About gobolinux

I can suspend my laptop and it restarts happily enough in X, but playing video results in a strange green square. I've found that starting and killing another X server (like X :1, wait for the grey mesh and then zap it) fixes the problem and I can switch back to my original :0 X and play video again. I wonder if these notes on X suspend and video BIOS by Matthew Garrett explain it.

About debian

Why on earth did someone change the .changes format so swiftly and why does a change that breaks a common upload process (build in a VM, sign on a stable system) only warrant a -devel-announce paragraph under the headline Misc Development News (#6)? "Small news" - my foot!

My communications connection

I'm going travelling this summer. I don't know what internet connection I'll have (if any) but I'm pretty sure I'll have fairly cheap SMS access from my phone. Maybe even MMS. So, I want to use one of the mobile micropublishers to try to avoid sending international SMSes to lots of people.

Facebook doesn't look good - it gives a shortcode (which would cost me extra to use and I don't know whether it works while roaming), has no number for SMS that I found and it looks like O2 is the only UK carrier it knows - bizarre.

LiveJournal has mobile access but the FAQ makes it look like it's only for paying users. If I was sure it was going to work, I might pay and support the GPL'd codebase.

Jaiku is cost-free and gives an SMS number, but is joining the Goolag and I don't see how to download their Java applet.

Twitter is cost-free, gives an SMS number and looks like it plays nice with Jabber (which is already on my phone), so that looks good but I found some complaints about whether SMSes get through.

20six is a German-based cost-free service which takes SMS and email (which is also on my phone), so that also looks good, but there are some old doubts about whether it will stay cost-free.

Is there a service you'd recommend?

Tux likes this fish-like object.

I finally got bluetooth file transfer working between my phone and laptop a little while ago. It wasn't particularly hard, although there were a couple of dead ends.

The basic bluetooth layer is petty easy. Start dbus, start hcid, start passkey-agent if you've not paired the two. It seems to be a bit simpler to start the pairing from the phone. Then use sdptool browse to check the phone is seen clearly.

After that, it got a bit complicated. obexftp worked well enough as far as it went, putting items onto the phone and getting files off the phone, but I couldn't work out how to get some items off the phone. They just didn't appear in the obexftp or obexfs listings. So I wondered if it night be easier to start the transfer from the phone.

There's an obexftpd, but I didn't figure out how to send it files. I also tried to compile opd but I think it offer patching for modern gccs. (Anyone got these working?)

What I did get to work was sobexsrv - just tell it a directory and it puts any files it's sent there. Works a treat.

One other thing that has been really useful is anyremote. It's a command server for the GNU/Linux side and a Java client for the phone. The phone mostly picks from option menus or preprogrammed keys, but editable fields are also possible. You can run any commands that the server configuration allows, including starting file transfers. I think that's possible because remote control and object transfer are on different bluetooth channels, but I don't really understand it yet.

I've improved the RSS reader to use xsltproc and added shell commands to it. Once I'm fairly sure it's reliable, I'll upload it near here.

(Posted in part due to an ALUG thread which reminded me about this forgotten draft.)

Is this traffic or congestion?

This is one of those Friday Afternoon Projects - it's been put off all through a busy week because it's unpaid, I'm not completely sure how to approach it and now my arms hurt like hell from travel jabs which are making it hard to concentrate! So I'm going to float it on here...

I've been asked to brief a meeting next week about that group's current web site and its problems. I'm not linking it yet to avoid insulting/embarrassing them.

The site looks OK, but doesn't rank well on search engines and doesn't allow much member participation. I need to explain why that's a bad thing and how the site's technical choices have led to that. I'm not directly pitching for TTLLP to get any work (because I'm a member of that group, it might be a conflict of interest and we're pretty busy anyway - even our own site needs work on some of the points I'm going to mention), but I don't want to be unhappy if we're asked to implement my recommendations.

I've got a usual outline that I follow, but my presentation's time is limited, so I'd like to ask you: what about this is important and what isn't? If you give me useful feedback, I'll put you in the Acknowledgements with a backlink and I hope the briefing will be shared pretty widely over the next few months.

The current plan is to start with a basic explanation of how search engines rank pages, as far as we can tell, referring to PageRank Explained Correctly with Examples, by Ian Rogers as well as the shorter official summaries from the dominant search sites at Yahoo, Microsoft and Google.

Then I go through a quick evaluation of the site against the basics of validation, accessibility and robot-friendliness, followed by a couple of SEO-style checks of its current rankings and inbound links.

Next is a bit different because I have access to some of their web access stats: I summarise what we know and suggest some other stats they've probably not considered and why they're useful, along the lines of Dave Briggs's measures of blog success.

Finally, I suggest ways to improve the site. The top tip will be to take control of the site hosting and stop using the cheap and cheerful donated server that makes all links except the front page point to another domain. I'll probably suggest a mix of free and open source software tools to power it. If they don't want to move it all yet, I'll suggest running a second site for member participation, using tools like Wordpress, NoseRub and so on.

What do you think? Plan for success, am I missing some tricks, or am I setting myself up for a lynching? Let me know with a comment or email, please.