Date blogger

Well I can see the problem with the date. The code I have introduced means that the date and month part must appear below the day abbreviation when it is a Thursday. This, in turn, means that the text doesn't flow correctly.
I'll have a look at what is most elegant to solve this.
I think, from inspection, that the abbreviation treatment is correct.
I have introduced a poll. It seems like a poll cannot extend beyond the month it is set in, which I find annoying.
I am still not happy with the use of the screen. I think the side panels take up too much rook, and waste space when there is nothing in them. The current design has two RHS panels, a narrow one and a wider one.
I also don't like the background to the gadgets.
What I do like is the way I am using MarkMail.
What i have done is narrowed down a search, in the case of the the one from scala-internals, to just two posts from David Pollak, to just 2.
Appart from the fact that this makes me look very cleaver if I can understand what he has posted about, this is genuinely very cool and useful.
What it is is a way of making comment on the list postings without having to post to the list. This is important to me because my comments will typically be off topic and, so, very annoying if made on the list.
I will elaborate on what I have to say about this post later.
In the meantime I also need to find a way of gathering data from MarkMail, if this is possible. It would be very interesting to have David Pollak has posted x number of emails on the subject of ... . Actually David has posted a huge number of mails, as it happens. I am unclear if this is a good example of what I am driving at or not.
I do remember that MS released a betta (it was quite broken from what I remember) many years ago of software that built a social picture of the interrelated posts on a topic or series of topics. I'm surprised this hasn't surfaced in any recognizable form, but, then, I haven't really been looking.
I notice that the blog doesn't display correctly as the blog entry is overwritten by the date. I wonder what the difference is with original template.
Well, obviously, the way to change all tags is to select them on each and every post where they are used, delete and then reinstate with the new spelling or format, in this case all lowercase. This might not work if there are a lot of posts!
On the other hand it is gratifying to notice that if a label is not used, i.e. the title case version, it does not appear in the show all labels, yes, tags are labels here, of course, and the philosophy behind them is a bit different to other tag usages.
I promised come back to this.
First of all I have decided to move all tags to lower case, but that is appro pro nothing apart from the last post.
The thread of thought is in the notion of insight into the evolving world of semantics.
First of all that is not my professional world, so I only know a little about it. Still there are ideas to explore which I shall come to over time.
Further I will be unselfconscious in the statements I make and the questions I ask.
I don't mind seeming foolish, no doubt I am foolish.
This applies to all of my ramblings. Call me a fool if you will, it's OK.
An example of my foolishness would be in the use of tags.
What, actually, are tags for and what do they have to do with web 2.0/3.0 - the semantic web?
You do not have to be in this industry for too long to come up against received opinion, the twist is that received opinion here has been formed about things that have existed for hardly any time, e.g. tags, but the way of forming those opinions is the same as ever.
So someone may argue the benefits of tags and tagging as if there is something in them more than what is quite apparent about them.
Tags will be sold, and I use the word deliberately, as a semantic idiom, something that expands meaning on the item that has been tagged, that is to say, enriches our understanding of that item.
Tags will be sold as a facility that fits in with web 3.0 and complements it.
But I think it is just that, a sales pitch given in a work situation where advocates don't want to lose face and want, if they can, to sell a facility for the benefit to themselves of having that to their credit, without thinking too hard about what, really, the benefit is.
I am speaking of a real work situation I have witnessed here.
But my rule of thumb comes in here, that is where there is one instance there is at least 2 * one order of magnitude other examples that lie beneath the surface.
So here I am talking about the pitch for tag enablement.
- My rule of thumb is not for the situation where the phenomenon is absolutely ubiquitous. An example of the latter would be where Ben Goldacre, in Is this a joke? noticed a report about the connection between crime, imprisonment and re-offending.
His conclusion was, to paraphrase, that the authors had not understood, or possibly even read, the reports from which they formed their own report and that the only people who could be interested in this report would be those who had no interest in the underlying arguments.
That behaviour, I would say, is ubiquitous and not subject to my 2 * one order of magnitude rule.
Coming back to tags.
Really this depends on how they are used. In the context of a blog they aford a useful way of sorting information and filtering, and of displaying the basis for filters.
But I don't think that is more than adding a single flat association to each item in the first place. I don't think that tags expand the meaning through description of the tagged item.
I also can't see how they can be reasoned over as they have little of no relationship to each other.
Still there is a frustration here. Why is this? delicious does a very good job of suggesting tags now. MT (MovableType) does an even better job, more later.
What with AdWords and so, I would have thought google would offer something in that direction. But I think that is a bit of the problem. The sort of tag suggestions I might want wouldn't be the same as what might appear from AdWords, and the priority would be entirely different.
Something to think about.
Well I had to go into the Javascript to create a conditional so that Thursday will appear as Thurs and all other days will appear as three characters (Sun, Mon etc.). No big deal, just detail, and a strategy to deal with the detail since I didn't know what I was doing.
Most of the time it was finding that strategy.
I have night mares about setting up a decent debug environment, my next task at work. No doubt the same principal.
I then had to find the setting for the time zone, it is a basic setting.
w.r.t the tag cloud I'm unsure how it will look when there is a proper differential in usage. But suspect I have seen better implementations so i may change it or use a different impl.
I have to decide a couple of things. Am I happy with capitalized tags?
If not can they be changed across the site, I suspect so.
Can I import my tags from Firefox and/or delicious. It is bad enough that i have these different collections of tags, in different formats. I gues next trivial task will be to see whether I can find (or create?) a conversion tool that harvests and publishes tags between different on line presences. This is a bit of a bizare business since there is also flickr and, at a stretch though I don't use either, linkedin and facebook and locally digicam and nepomuk. There is also Last.fm.
I think that argues well for some uniform tool and there have been efforts in this direction. One was a DB that was built out of harvesting publically available tags, obviously in the line of what I am thinking. I think that service has, more or less, closed or sells into advertising agencies. This last is a sensible move and might be worth my investigating again?
About tags - well this is a whole other post, but just entering them
  1. It is far less effort to enter in lowercase.
  2. I am unsure how to deal with plurals - do I mean services or service? What about blogging, blogger and blog which are two nouns (blogger, blog), one with two meanings, blogger this service, blogger a person and two verbs (blogging, blog). Em.
  3. From my experiance with delicious, how to deal with compound words or concepts.
  4. How many tags to apply to a post anyway - that is how to know that more is not better? It just seems that fewer is easier for the user, but that depends on usage ...
Well, I have messed around with the template and feel happy enough with it.
Some points.
This is a note pad. I'm not too concerned if anyone reads this.
This blog is oriented to my work and carries my signature, so it has some restrictions on what I might say. This, in turn, probably is to the benefit of the reader, or at least the type of reader who may end up here, since, in some parallel universe there may be another blog, possibly posted to under an alias, that contains all my most creative imaginings and inconsistent ramblings, right down to the most acerbic, biting critique. But enough of that, that is not for here. That only may ever take place. And, anyway, when it comes to this industry I will make a critique, it will be suitably anonymised here to prevent red faces, my own or others, in anger or shame.
This brings me to the last point I wanted to make in this first 'real' post.
The blog is called SemanticC, somehow a cleaver name, but it really means next to nothing.
I guess it could mean insight into the evolving world of semantics. Perhaps it will turn out that way, I have no objection. I will elaborate on this point in my next post.
But looking at the first post from April last year says something about me.
A debate graph - what a good idea.
What it says about me is that I expect, anticipate, even like contention.
I could say more about this, the person who puts himself in the middle of opposing arguments to see if they can be resolved.
I will leave this point dangling, but it is a good thing to bear in mind when considering AI and semantics.
top