A storm blew in during the night with lots of rain. I always sleep with an open window an this morning I had a fair (well ugly really) bit of water in the bedroom.

I played around a bit with the XML/FTP programs (located an old bug 1998 vintage) and changed a few details in v7.9.
And then spent time on the documentation. Well I actually completed that but of course documentation, just like the programs is never really 'done'. Rereading it I picked out a bunch of errors (linguistic, not factual) and noticed a few incomplete passages. 't will be ready by tomorrow noon, just in time for the meeting.
Now before any of the writer among the daynotes gang makes a remark about me taken so much time out for documenting a program, hardly reaching a few thousand lines, that I must know intimately, I wrote it myself just jet last week, let me tell you it is not that simple. Not necessary simpler than writing documentation on a totally unknown program.
You see, when you write about an unknown program you can only write about what you see. You can write how to use the program from the point of an end user, albeit an end user with lots of experience. There are probably whole sections of the capabilities that you don't ever notice. Not least because a lot of them aren't intended for end users.
I wrote this stuff so I know all the possibilities. Most of them don't show in the interface. So should I put them down? Of course. Because this documentation is not just intended for end users but also for system administrators that have to fit the programs into their daily (or monthly or ...) operations (all information needed to use it in a script must be there). And it's intended for programmers that may have to incorporate it into their own processes (that means all backdoors must be mentioned).
And of course it all must be written in such a way that managers can understand it. That means short (if it is more than 2 pages (excluding screen-shots) they don't bother to even looking at it. And it must sound impressive, well when they explain it to their clients/bosses it must.
The biggest danger though is that I write down what I intended the program to do and not what it actually does.

It's only after doing such documentation jobs that one really appreciates the work an author has to do.

The problem now is what to do next. There is more than enough work but the priorities are not clear. Most things on my to-do lists are flagged with high priority.

The temperature is getting up a bit (8C this morning) with just a tad of wind and some drizzle that make for real uplifting weather :(
It's just that I don't like this undecided type of weather. I much prefer something like what Tom has over on his spot of the world. Not that I would mind to swap weather with Dan or Brian . Anything is better than this yes-no type of stuff.

While working on the documentation I hit a few problem in the program. Minor stuff really (like not passing the sign on dimensions, only a problem if an object with for example a negative weight is catalogued) but then again I like to produce bug and problem free software.
So I adapted the programs (and documentation). And I found an old problem (ano 1997 at least) in the FTP setup program. To be hanged solved tomorrow at dawn.

The meeting about the character sets to be used in our XML files (the solved problem) turned out to go a bit deeper than I thought. Basically nobody ever thought to work out a consistent method of development internally so that now whenever two modules or programs have to communicate the character set is decided fresh. The result is that Robby has to work with different conversions each time he moves data from one spot to another. So no only the current (solved) problem was thrown on the table but a bunch of other stuff landed there as well.
After about ten minutes I knew the final decision (keep what we have and look into Unicode (ucs-8 ) for new developments). But of course with three manager involved ten minutes would never do. One hour was scheduled so it had to be stretched out a bit, to a total of about one and a half hour. To indeed reach the afore mentioned conclusion. I would have closed the meeting after at most 15 minutes (10 minutes discussion and five minutes to finish off the coffee) but that only shows that I have no management skills.
It was fun though to see them do it.

 

Spammers at work (well 'work' ...).
I notice that spam mail isn't coming once anymore. More and more spamers seem to be seeding different spam mailing lists resulting in the same spam dropping in the mailbox just seconds apart.
I got a kind of negative filter running now. Anything coming from certain non-spamming sources (like Daynotes Mailing List) goes to a specific mail folders. Anything explicitly sent to me filters into my active folder and all the dross remains in my inbox waiting patiently for incineration.
BTW my dictionary doesn't know spam and proposes 'spasm'. quite rightly so.

The weather varies between, on the one hand rain and drizzle and on the other hand almost rain and drizzle. No sun to be seen. With things being overcast it doesn't cool down at night of course (8C) neither does it warm up significantly by day (11C).

The XML stuff is done, including a 'last minute' change. With the documentation packed up and sent off to the boss, I mean The Boss.

So I hanged solved the FTP problem. Rather easy, just use another logical (index) and pass an extra parameter around. Typical 15 minutes coding, an hour testing and an hour shuffling programs from one environment to another.
After killing the beast I start to pull things together for writing up some useful documentation on the FTP interface. Maybe I even finish that sometime tomorrow.

In between I get to solve a problem in the Quality control programs. I solved a serious, crashing problem just last week. After solving and testing I was not sure I had things working correctly. There were no more crashes so that bug got crushed but I didn't have the necessary authorities to actually test all the changed code. Driving home I got the insight that I had forgotten to initialises some fields but in the morning the insight had been pushed into some of the many corners that riddle my head. Now Pieter had tested the program and of course he did see the problem. Immediately after he notified me the insight left its corner and slowly walked up to centre stage. So I solved that fast enough and this time I could test it fully (by breaking the authority rules).
BTW Pieter is a real nit picker but at least he does his testing seriously and thoroughly. Unlike Jan.

I am getting used to this. Overcast and grey with lots of short showers and two (exactly) outbreaks of sun.

Did I do something? Yes indeed, I was quite busy even.
Ha but you want to know what I actually did?
Heu ... Well, ... now see ...
Yes.

Nothing memorable happened. I did a bit documentation and I shifted a few programs around.

Overcast and grey with some short showers and two (exactly) outbreaks of sun. Any one notice the difference? Well the big difference was that the second outbreak of sun started at about 15:00 and ended at sunset. So not two times five minutes.

I did mostly the same thing as yesterday. That is tying down loose ends and wrapping up open projects.
Most of the time Ronny or me are the first in the office (after Frank when he is in) at around 7:40 with Theo coming in at about 8:10. Today Theo was in first. I did look at the clock because my first thought was that I had missed something but no it was 7:40. Turns out he had to prepare a demo of the Catalog and the XML/FTP stuff for the Germans. I did a quick check but all programs were moved over to the right machine.

Because he is being overloaded he just gave his approval to start with the order-validation adaptations that we discussed last week. Without waiting for and consulting with the opco's. Ha, but do they actually want it? No, ha yes, well maybe. It will solve some requests, actually going a lot further. I call it anticipating, going for the best solution instead of improving the current situation step by step.
The only problem is that we don't have a written specification. The only written stuff is the original proposal that was used as a starting point for the discussions. All the rest hangs in thin air and memory. So I started with condensing all that thin air into a document-like structure.
Now, a 17 inch screen at 1024x768 isn't big enough to put two documents side by side and I wanted to see the old document (MSWord) while I was composing the new one (CuteHTML). So I pulled out my notebook, copied the Word document on it and opened it in Star Office (it is my personal notebook so it obviously doesn't have MSWorddt on it :). I know a 17 inch and a 12 inch display doesn't combine to a 29 inch screen but it was workable.

BTW, I am using Lotus Word Pro, and that wants to replace MSWord with Sword. Maybe because of the sharp edges?

I told you yesterday that we had some sun yesterday, due to an acute lack of clouds. That lasted the whole night pulling the temperature down nicely. To about 1C. Later the clouds returned and the temp got up again. The day itself was again cloudy and windy but dry.

The low night temperature does mean that we should start getting some plants inside against the frost. Normally my father does that but I get a feeling he is not going to start in time this year. Maybe I should start.

Shopping. Well, I went into town but of course all shops were closed. Armistice Day is a public holiday here.
A public holiday? Yep so I make it into a complete rest day.
















Halloo, still hanging around here?
Didn't I make myself clear. A rest day.

Dry and cool (7C) with a brisk wind in the morning turns into a drab wet day. The rain constantly varying in intensity but never letting off.

Reasonable running weather with a good cooling wind.

I don't remember doing or eating anything wrong. Yet I got internal allocation errors. Illegal operations of the biological kind, only solvable on the toilet. So that is where I spend an unusual proportion of my time today.
Not a condition for productive endeavour. I just picked up a book (The Underpeople from Cordwainer Smith) and read that trough. Not much of a story there.

Later I worked a bit on my wine pages. I still have to come up with a background though.