We've a Notes client application that contains a form with an Edit/Read Document toggle button on it, that's supposed to switch the document into edit mode from read mode or vice-versa. It was coded to use @command([EditDocument] to do that, and mostly worked OK.
We had sporadic reports that when it was clicked, and entirely different document was opened. And recently, we've been able to track down a repeatable scenario to make it happen:
- Do a Full-text search in the default view
- select a document from the result set, edit it and make changes
- go back to the result set and clear the changes
- position the cursor on another document
- go back to the open document
- hit the Edit/Read toggle to set the document back to read mode
And instead of the status change on the current document, it opened a new one. Not the one which was toggled, and not the one selected/under the highlight bar in the open view either. By the way, the user scenario was leaving the database open when doing something else, and then coming back to it and starting a a new task, whuch was why it took so long to track down, because when it was reported, the user had no real idea of the set of steps required to repeat it.
As you might imagine, we were somewhat perplexed with this. We changed the button code to LotusScript:
Dim ws As New NotesUIWorkspace
Dim uiDoc As NotesUIDocument
Set uidoc = ws.CurrentDocument
uidoc.EditMode=False
which showed exactly the same results. We ran the debugger, and uidoc was the currently open document, but the uidoc.editmode command still opened a new document, apparently randomly.
We've fixed it, by altering the code to grab the UNID of the open UIDoc, close it, and then open the target document by UNID, but we're still puzzled as to why it happened in the first place.
Comments (2)
Mick Moignard November 20th, 2008 04:39:18 AM
Adding Blackstone drawbars to brass locos isn't particularly hard, and makes servicing of them easier because you can separate tender from loco when needed. Heres some notes written specifically to cover Nakamura (Westside, PSC 1984 run) and MSM (OMI, PSC 2002 run) models. I've used similar techniques for Nakamura (Westside) K-27s, PSC/MSM K-27s, PSC/MSM K-37s, Fuji K-37s and both Nakamura (Westside) and PSC/MSM K-28s, all in HOn3.
Nakamura K-36s: These are easy. Remove the loco drawbar and the pin from the tender (it unscrews). The hole in the tender is tapped M1.6 size; you need 2 M1.6 bolts and one nut. The hole in the BS drawbar is around 2.5mm diameter, so you also need some plastic tube - I think I used Evergreen 1/8th inch tube reducing the outer diameter carefully to 2.5mm with a file, and then boring the centre (hold a drill in your hand) to make a tight push-fit on the tube.
I removed the bottom piece of the BS drawbar on each end, and opened the hole out a little - this allows the drawbar to tilt up and down a little. BS do it on the K-27s by mounting the drawbar on a tapered pin; I just enlarged the bottom hole instead by about 10-15thou.
Mount the tender piece using a piece of the tube around 5mm long and a suitable length bolt, so that the bolt cinches up against the tube, and the drawbar half can pivot freely side to side and up and down. I use the female part on the tender.
Similarly mount the loco piece with another bolt and the nut,, using the existing plastic filler piece to put the bolt thru. You need to experiment with the tube length a little so that you get the drawbar at the same height as the tender end. The loco and render will be slightly closer together than they were on the Westdide drawbar, but are good for 24-inch curves with no shorts.
MSM K-36. These are a bit harder; you could just use the standard holes for the loco and tender drawbars, but then the tender is too far away from the loco and looks silly. The tender takes a little work, in both cases. Remove the pin. File the pad that the pin goes in to flat and level with the bottom of the front tender sill. Then drill a new hole around 1.7 to 1.7mm further back, and tap M1.6, or whatever size bolt you plan to use. Then prepare the drawbar and plastic tube pretty much as for the Westside.
On the OMI loco, you can just mount the loco end of the drawbar as per the Westside, using am M1.6 bolt and nut. You may need to bend the ashpan rod that runs side to side across the loco with a kink forwards in it a little further forward to clear the drawbar. This rod represents part of the linkage that opens the ashpan doors for dumping the ash.
On the PSC loco, you have to remove this- it's a casting here rather than a piece of rod. It's central part gets in the way of the drawbar, so you can remove all of it - unsolder at both ends and pull out, or clip the middle out, or do as I did and remake it with a larger space between the two outer forward extensions. Your choice. I remade it with new brass rod, clipping out the two outer levers and the two inner ones and drilling them all for the new pieces of rod.
Then you can mount the loco drawbar as before with an M1.6 bolt.
Drawbar wiring: I always wire the drawbar the same for every loco, so that I know which wire is which. Using the two outer ones enables me to be able to spin the motor if need be with clip leads. The female part is always on the tender, so that there is no risk of shorting bare decoder connections should the drawbar come open. Run the left four wires one side of the mounting bolt, and the other four the other side.
| Drawbar wire colour | Decoder wire colour | Used for |
| green | orange | motor - usually the positive terminal |
| blue-yellow | red | right rail |
| blue | blue | function +ve |
| red-green | tan (big Tsunami) | cam wire. Brown on the small Tsunami. |
| red-yellow | white | headlight/F0F |
| red-blue | green or brown (big Tsunami) | Functions FX5 or FX6 as appropriate |
| orange | black | left rail |
| red | grey | motor, usually -ve terminal |
Comments (0)
Mick Moignard November 18th, 2008 01:35:00 PM
Bill's comments on the hedge funds that have lost out when shorting VW shares just about says it all. Chances are strong that these idiots were borrowing shares from Porsche to sell them to Porsche and then having to buy them back from Porsche (note that Porsche returned 5% of VWs shares to the market at one point) so that they could give them back to Porsche.
I also noted a report in BBC Ceefax a few days ago that said that the Tokyo exchange had banned anyone from selling shares that they don't own and had not borrowed in an attempt to remove speculators from the markets. Given the way that markets have been going recently, it look as though speculation is still rife, and generating serious instability.
Why don't the markets just ban such activity altogether, and require that you cannot sell shares unless you have clear title to them? And let the stock markets go back to the business of providing investment finance to business, rather than acting as a vehicle for extreme gambling?
Mick Moignard October 30th, 2008 08:44:37 AM
EditorDave has just published an article in DominoPower with some insights on how judges think about Lotus Awards - these are what used to be called Beacon awards, and are given at Lotusphere each year. I guess it's no real secret that I was one of the judges for last year's Composite Apps award, and Dave has used some input from me, along with feedback from half a dozen others in writing the piece. I won't repeat any of the advice given here, because then you wouldn't read the article.
I did enjoy the process last year, reading and evaluating the submissions. In my category we had quite a few, with some really impressive offerings, even if some of them appeared to be submitted for the wrong award. Given that we had a couple of dozen or more submissions, and each one deserved initially a few minutes to understand and categorise, plus the fact that the serious contenders then needed up to an hour or more to review properly, it took quite some time to whittle my views down to a manageable and discussable number. The judging team then had a number of conference calls to finalise views and declare a winner. Certainly every one of the submissions I looked at showcased quite some technical achievement, even if it wasn't always well described or one that fitted the classification, and they showed off a wealth of talent in the Business Partner community.
If you are thinking that 2009 might be your year for an award, start to think now - if you haven't already - what you could win at. Make sure that you know what makes your solution stand out in your chosen category, and make sure you know how to tell me and/or the other judges why you think it it stands out and why your solution is the best one on offer. You need to grab our attention with what you have, and make sure you keep our attention.
So, read Dave's piece, and make sure you read the submission rules and category definitions carefully. Good Luck for 2009.
Mick Moignard October 22nd, 2008 10:05:46 AM
One of the things that hasn't helped things over the last few week's turmoil in the financial world is the short-selling of shares; a practice that certainly here in London is supposed to have been stopped, and maybe it has - though there seem to be some getouts and not a lot of sanction. But I've been puzzled over how it actually works:
Company A who owns some shares; probably an insurance company, agrees to lend them to B for an agreed fee.
B, who thinks that the price is going down, sells them.
B, after a while (a few days or weeks) buys them back, at hopefully a lower price.
B then returns them to A, pays the agreed fee, and pockets the rest of the profit.
To make this work well, it's in B's interest to try to depress the price of the shares as much as possible, by I guess what ever means he can get away with.
Now, here's the bit I don't understand. B gives A back the shares he's borrowed, but those shares are worth less than they were when A lent then to B. Even the fee B pays for the loan can't cover that, else B would not have made any profit. So what's A (who, by the way may well be managing my pension!) doing lending them to B in the first place, knowing that they'll be worth less when they come back?
Mick Moignard October 14th, 2008 10:00:25 AM
From the log.nsf
03/10/2008 16:02:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:03:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:04:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:05:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:06:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:07:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:08:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:09:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:10:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
03/10/2008 16:11:13 Process C:\Lotus\Domino\nserver.exe (4300/0x10CC) has terminated abnormally
Been doing it all day. Still works OK as far as I can tell, though. Restarted it, and it's still doing it. Once every minute. Strange?
Comments (1)
Mick Moignard October 3rd, 2008 10:55:57 AM
One of my co-workers came upon something curious today. She was looking at the properties of a document from a view, and noted that quite a few fields weren't shown in the properties box. Several eliminations then ensued:
- Yes, Allow more Fields In database was enabled
- No, the field(s) in question are not computed for display
- Yes, the field is on the form - not that this should make any difference
We did note that the document had approx 3000 fields on it, and concluded that the properties box is somehow challenged; it cannot show all the fields for a document that has a large number. Which begs the questions:
- what criteria does it choose the fields it shows? it appears to drop fields with higher SeqIDs
- what number of fields can it display up to?
- why is the properties box able to show all the fields when the document is actually open? Is the issue not the properties box, but maybe something to do with accessing the fields when the document is not open; does the properties box open the document "differently" when the document isn't open in the UI?
It's not client version dependant either; we can see the issue with both 7.0.x clients and 8.0.2. Comments (7)
Mick Moignard October 1st, 2008 05:45:01 AM
I installed Hotfix 19 for Quickr 8.1 for Domino yesterday, on a test server, and then reinstalled the connectors on my laptop. And Lo! Connectors installed in my Notes 8.0.2 sidebar, mail file and even a link in the Sametime sidebar - and they seem to work; the Quickr Sidebar can even see and interact with not just the Domino Quickr 8.1 but also a Websphere Quickr 8.0 that we also have on test.
I have to say that I was half-expecting it, because I'd remembered seeing it mentioned in a fix list somewhere; turned out that the list I'd seen was in k-Base announcing Q 8.1.0.1: technote 7013341. At the bottom it includes this table of connector fixes. So I wonder whether these connector fixes are actually in HF19, or earlier. Anyway, they do appear to work with 8.0.2, and that's good.
| Defect ID | Build # | Description |
| LO28175 | 8.1.0.92 | Cumulative Connector fix, which addresses Outlook Crash and Upgrade install failure (supersedes APAR LO27612) |
| 207117 | 8.1.0.95 | Office: prompt window does not stay on top |
| 206314 | 8.1.0.95 | Outlook: Sort order of library list on Taskpane |
| 206332 | 8.1.0.95 | Outlook 2003: Error opening unsent msg files (MS word editor) |
| 206933 | 8.1.0.95 | Office: save to Place As, wrong toolbar status |
| 206388 | 8.1.0.95 | DnD email contain attachment and title end with space issue |
| 200167 | 8.1.0.95 | DnD a mail containing " in subject does not work |
| 208531 MMOI7DHK2U MMOI7DKPX9 | 8.1.0.95 | Policy to disable Remember this password option not honored in Quickr html link; Policy to disable Remember this password option not honored for multiple NT users |
| 208329 | 8.1.0.95 | Upgrade from connectors 8002 to connectors 8.1 |
| 210857 | 8.1.0.101 | Quickr 8.1- need to enable install sidebar in Notes 802 and 8.5 |
| THES7DRQP8 | 8.1.0.101 | Can't disable Quickr Icon on desktop |
| YXIO7E9563 | 8.1.0.102 | Notes crashes when folder/file path too long |
Comments (3)
Mick Moignard October 1st, 2008 02:27:41 AM
No.
I've been rather amused by all the hype today, with the switch-on of the LHC at Cern. It was never going to destroy the world today, because they haven't actually got that far yet. As with all large machines, commissioning is a complex process, and isn't just a matter of flicking the power switch and seeing what happens.
All they've done today is started up one of the two rings, sector by sector, and have shown that they can run hadrons around the ring in one direction. Note that there are actually two rings, which operate in different directions, intersecting only at the detectors; it's only at the detectors that the to beams intersect and collide, and only then if they are set to collide. They do intend to start the second ring today, but they don't plan starting any collisions until October. That's when it starts to get lively - and according to Cern themselves, microscopic black holes are not the only things that have been speculated about. But even if they did generate one, it would be so small that it would likely be centuries if not millennia before it swallowed the earth, so we'd have time to do something about it.
Bit of course this stuff is actually rather old hat. Cern's LHC may be that largest and most powerful collider around, but it's by no means unique - the next largest is the Tevatron at Fermilab in Illinois, and there's no reports that it has generated any microscopic black holes in the 20 years that it's been running. And the x-ray machine used by your dentist works basically the same way (firing a stream of electrons at a beryllium target, and x-rays come off). No black holes in my teeth, as far as I know.
Mick Moignard September 10th, 2008 06:50:23 AM
I've just spent a day exploring Mashups, and one of the things I've done is follow one of the tutorials through. At the final save, got this: (Read it carefully; I didn't spot the problem at first, either).
Needless to say, the thing doesn't work, but I don't know why.
Update: If you use IE7 instead of Safari, this message doesn't appear. But developing a Mashup with IE7 is not a good experience because, compared witH Safari, it's desperately slow: To load the feed mashup generated in one of the tutorials takes 4 seconds with Safari, and 24 seconds with IE7 - same Mashup server, located on my laptop. However, Safari then shows the following error:
Hmm.
Mick Moignard September 9th, 2008 03:29:05 AM