test please delete

[voodooflux]

where has everyone gone ?

[zara_backdog]

just downing some games for the kidss

[voodooflux]

just downing some games for the kidss

I can empathise - sometimes I feel like the technical support dept. in the Flux household.

[zara_backdog]

I can empathise - sometimes I feel like the technical support dept. in the Flux household.

I would not mind I am still trying to get use to the mouse

Not good after a bottle of wine and a new laptop Hic

[voodooflux]

I would not mind I am still trying to get use to the mouse

Not good after a bottle of wine and a new laptop Hic

I have enough trouble when I'm sober

[voodooflux]

Time for bed

[NickFitz]

where has everyone gone ?

To the pub, to celebrate the new gig

I see that has seated him/herself conveniently close to the <textarea> element in which one posts, but YMMV

We should organise a test sometime to see if everybody gets the smilies in the same order

Although one has to bear in mind that the last time we tried to organise anything on TPD was giving Carling Wotsit the 21KPost - the merry mayhem was most entertaining to watch, as the denizens initially went into an uncontrolled frenzy of post deletion to give Carling her 21K, which then resulted in too many posts being deleted, until finally stasis was achieved - until the 100K post deletion, and then the random deletion the other month (around 130K, IIRC?)

Idiots! We've Gone Too Far!!!!!!!!!!

I'm still trying to derive an algorithm that copes with all this. Given situations where, for example, a post back in the 3xxx area is deleted, a binary search is clearly good: but remember that each floor(partition mod 10) involves an HTTP GET, so it has to be an asynchronous binary search - and then we run into the interesting case where a recent post is deleted (I'm looking at you, Churchill... ) and a newbie post is then approved while the "detect deleted post" algorithm is, asychronously, chuntering on; for suddenly the hole would have been filled and the upper boundary of the search would have altered due to an asynchronous change in the underlying dataset

This may, or may not, be one of the Great Unsolved Problems of "Computer Science". Either way, I look forward to either solving it, proving that it can't be solved, or just becoming tremendously confused and going to the pub instead

[NickFitz]

Speaking of systems engineering confusion, the traffic lights at the junction here seemed to have lost the plot when I was returning from the pub tonight.

When I arrived at them (as a pedestrian, obviously) they were red for both aspects of the crossroad, so I waited, there being a couple of vehicles also waiting on the main road.

After about thirty seconds in the wind and rain, I realised that they were never going to change - they'd fallen into some state of fail where, to be safe, they were remaining on red for all directions

So I hit the button on the "awaiting pedestrian" box. This input seemed to kick the whole system back into gear: the lights immediately turned green for the main road (not the intuitive response to pedestrian input, but I assume the system was doing what it had been programmed to do, and recovery from failure has to start somewhere) and, after the awaiting vehicles had cleared the junction, I crossed. Looking back, I saw that the system correctly moved on to the next phase of turning the main road lights red and the crossing road lights green.

The final confirmation that everything was hunky-dory would have been to wait for the main road to go through another green cycle and then (assuming that the system recovery hadn't wiped out the pedestrian request) switch into "pedestrian crossing" mode, for that comes after the main road green at this particular junction. But I wasn't going to hang around for that - the wind and rain are bitter

I believe we're entitled to snow tomorrow

[BrowneIssue]

I got woken by a bad dream of a man pushing his way into the house and refusing to leave.

[BrowneIssue]

I lay there wide awake afterwards wondering what had caused that.