A collaboration with Domino developers about how to do it and how to get it right in Dominohttp://www-10.lotus.com/ldd/bpmpblog.nsf/en-usFri, 11 Jul 2008 04:19:25 PM -0400 Fri, 11 Jul 2008 04:19:25 PM -0400 Joseph LeMay Creating documents in the future http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/creating-documents-in-the-future?opendocument&comments#07112008041925PMWEBRU7.htm Mon, 7 Jul 2008 03:25:49 PM -0400 Gabriel Amorim Thoughts while waiting for my meeting to start
[...]

Still I do not love Comic Sans...]]>
[...]

Still I do not love Comic Sans...]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/thoughts-while-waiting-for-my-meeting-to-start?opendocument&comments#07072008152549WEBQSP.htm Mon, 7 Jul 2008 01:49:45 PM -0400 Tripp Black Thoughts while waiting for my meeting to start
As for the default vs model, I like the default to be "Snooze" or "OK" -- I would tend to OK more than snooze but can get to used to either way. That way I can just click and get back to work.]]>
As for the default vs model, I like the default to be "Snooze" or "OK" -- I would tend to OK more than snooze but can get to used to either way. That way I can just click and get back to work.]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/thoughts-while-waiting-for-my-meeting-to-start?opendocument&comments#07072008014945PMWEBNWN.htm Mon, 7 Jul 2008 11:39:25 AM -0400 Charles Robinson Thoughts while waiting for my meeting to start
@1 - I agree, that is a nice touch. One thing that bugs me is the Outlook window doesn't have an "OK" or "Done", you have to click "Dismiss". That's not at all intuitive to me. Every time I see the dialog I'm looking for "OK" and it's not there.]]>
@1 - I agree, that is a nice touch. One thing that bugs me is the Outlook window doesn't have an "OK" or "Done", you have to click "Dismiss". That's not at all intuitive to me. Every time I see the dialog I'm looking for "OK" and it's not there.]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/thoughts-while-waiting-for-my-meeting-to-start?opendocument&comments#07072008113925AMWEBLD7.htm Mon, 7 Jul 2008 10:42:01 AM -0400 Brad Thoughts while waiting for my meeting to start
And as far as the dialog getting dismissed, that is something that is many versions late. It should be a modal window that does not have a default button so that I cannot accidentally dismiss it.]]>
And as far as the dialog getting dismissed, that is something that is many versions late. It should be a modal window that does not have a default button so that I cannot accidentally dismiss it.]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/thoughts-while-waiting-for-my-meeting-to-start?opendocument&comments#07072008104201AMWEBK9B.htm Mon, 7 Jul 2008 09:42:22 AM -0400 Martin Thoughts while waiting for my meeting to start http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/thoughts-while-waiting-for-my-meeting-to-start?opendocument&comments#07.07.2008094222WEBJ42.htm Wed, 2 Jul 2008 11:02:26 PM -0400 Erik Brooks The mysterious disappearing blog entry
{ Link }]]>
{ Link }]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/the-mysterious-disappearing-blog-entry?opendocument&comments#02.07.2008230226WEB5L3.htm Fri, 27 Jun 2008 07:00:33 AM -0400 Jesper Kiaer Creating documents in the future
Thanks for the response!

I agree it is very difficult to foresee what may happen in the future so sometimes decisions made, may some same odd or wrong later in time.

I do not think the is the case here.

A UNID sole purpose should be, to be unique. The creation of the UNID should not be dependant on "some information" should be extracted from it, like the creation date/time.

The idea that that only one document may be created in "a time tick", what ever that size may be, is wrong.

My customer experienced this the hard way on 90 servers worldwide. Setting a db property (unread marks I think) caused 100.000 of documents to be modified and time crept a week in the future completely messing up replications etc. Lotus Notes/Domino was not a popular platform the following weeks...

Hopefully it will be fixed in future a release.

brgds Jesper Kiaer (This)

{ Link }]]>
Thanks for the response!

I agree it is very difficult to foresee what may happen in the future so sometimes decisions made, may some same odd or wrong later in time.

I do not think the is the case here.

A UNID sole purpose should be, to be unique. The creation of the UNID should not be dependant on "some information" should be extracted from it, like the creation date/time.

The idea that that only one document may be created in "a time tick", what ever that size may be, is wrong.

My customer experienced this the hard way on 90 servers worldwide. Setting a db property (unread marks I think) caused 100.000 of documents to be modified and time crept a week in the future completely messing up replications etc. Lotus Notes/Domino was not a popular platform the following weeks...

Hopefully it will be fixed in future a release.

brgds Jesper Kiaer (This)

{ Link }]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/creating-documents-in-the-future?opendocument&comments#06272008070033AMWEBEWW.htm Thu, 26 Jun 2008 03:02:05 PM -0400 Gabriel Amorim The mysterious disappearing blog entry
LOL]]>
LOL]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/the-mysterious-disappearing-blog-entry?opendocument&comments#26062008150205WEBQBU.htm Thu, 26 Jun 2008 12:27:52 PM -0400 Andrew Magerman Diatribe: Macro formulas, temp variables for each @Function argument
I exactly have that. Long ago (this was before the nifty Hello:=0 as arguments), I made a collection of Teamstudio snippets with exactly those pre-formatted entries - no values, paste in whole code, adjust the necessary variable, done. No half-an-hour trying to work out that you got the third and fourth argument wrong. I especially alwayays got the long ones wrong, like @dblookup, @Adjust - impossible to keep those in head.

Andre, I know also that the formula language has been "proceduralized", but I still think in terms of a stack, and I think that still is the best metaphor for really understanding @Formula - and your comments seem to really assume a "start at the top and end at the bottom" logic.

Sorry, not convinced on this one, Andre!

Andrew

P.S. Apart from that, brillant blog!]]>
I exactly have that. Long ago (this was before the nifty Hello:=0 as arguments), I made a collection of Teamstudio snippets with exactly those pre-formatted entries - no values, paste in whole code, adjust the necessary variable, done. No half-an-hour trying to work out that you got the third and fourth argument wrong. I especially alwayays got the long ones wrong, like @dblookup, @Adjust - impossible to keep those in head.

Andre, I know also that the formula language has been "proceduralized", but I still think in terms of a stack, and I think that still is the best metaphor for really understanding @Formula - and your comments seem to really assume a "start at the top and end at the bottom" logic.

Sorry, not convinced on this one, Andre!

Andrew

P.S. Apart from that, brillant blog!]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/macro-formulas-temporary-variables-for-each-function-argument?opendocument&comments#26062008122752WEBMBG.htm Thu, 26 Jun 2008 11:30:16 AM -0400 Eric Romo Diatribe: Macro formulas, temp variables for each @Function argument
Sure you can do alot with formula language, but inherently its not well suited for complex programming. Thus keeping routines fairly short/simple, and using variable names are only very rarely a waste of time or cpu cycles.

IMHO, the original recommendations should be limited to when performance needs to be improved.]]>
Sure you can do alot with formula language, but inherently its not well suited for complex programming. Thus keeping routines fairly short/simple, and using variable names are only very rarely a waste of time or cpu cycles.

IMHO, the original recommendations should be limited to when performance needs to be improved.]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/macro-formulas-temporary-variables-for-each-function-argument?opendocument&comments#06262008113016AMWEBL7G.htm Thu, 26 Jun 2008 10:11:56 AM -0500 Andre Guirard Creating documents in the future Why on earth has this "unwise" way of creating UNID not been fixed long a go?

Dear This,

I haven't talked to the the people who are looking at this, but I'd guess it's a tough problem. This UNID is the created date/time in the current architecture; I don't think we store the created date/time separately. We represent date/time values in a certain way which is based on .01 second intervals. There's no other really good way to generate values that have the desirable properties of being translatable to a date/time which is also the created date/time, being in order for a binary sort, and being guaranteed unique in that replica once we do the clock adjustment for creating multiple in the same tick (we go for uniqueness across replicas by combining it with the database ID, which isn't visible in the UI).

I think in large part, the problem is that the original designers of the Notes architecture, and of the Windows OS and hardware, didn't really take into account how fast computers were going to get. The clock on Windows PCs ticks over at a certain rate -- can't make it go in smaller increments. We've got gazillions of date/time values already stored. If we tried to store them at a greater precision, we'd need more bits to store them in, or we run out on the high end.]]> Why on earth has this "unwise" way of creating UNID not been fixed long a go?

Dear This,

I haven't talked to the the people who are looking at this, but I'd guess it's a tough problem. This UNID is the created date/time in the current architecture; I don't think we store the created date/time separately. We represent date/time values in a certain way which is based on .01 second intervals. There's no other really good way to generate values that have the desirable properties of being translatable to a date/time which is also the created date/time, being in order for a binary sort, and being guaranteed unique in that replica once we do the clock adjustment for creating multiple in the same tick (we go for uniqueness across replicas by combining it with the database ID, which isn't visible in the UI).

I think in large part, the problem is that the original designers of the Notes architecture, and of the Windows OS and hardware, didn't really take into account how fast computers were going to get. The clock on Windows PCs ticks over at a certain rate -- can't make it go in smaller increments. We've got gazillions of date/time values already stored. If we tried to store them at a greater precision, we'd need more bits to store them in, or we run out on the high end.]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/creating-documents-in-the-future?opendocument&comments#06262008101156AMAGUKT9.htm Thu, 26 Jun 2008 08:18:59 AM -0400 Gerald Mengisen The mysterious disappearing blog entry http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/the-mysterious-disappearing-blog-entry?opendocument&comments#06262008081859AMWEBGFW.htm Thu, 26 Jun 2008 03:47:28 AM -0400 This "feature" has cost one of my customers lots of troubles Creating documents in the future
Why on earth has this "unwise" way of creating UNID not been fixed long a go?

brgds Jesper Kiaer]]>
Why on earth has this "unwise" way of creating UNID not been fixed long a go?

brgds Jesper Kiaer]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/creating-documents-in-the-future?opendocument&comments#06262008034728AMWEBB68.htm Wed, 25 Jun 2008 09:18:40 PM -0400 Nathan T. Freeman Diatribe: Macro formulas, temp variables for each @Function argument
('cause it's all about me, dontchaknow)]]>
('cause it's all about me, dontchaknow)]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/macro-formulas-temporary-variables-for-each-function-argument?opendocument&comments#06252008091840PMWEB3K8.htm Wed, 25 Jun 2008 05:57:18 PM -0400 Training? Diatribe: Macro formulas, temp variables for each @Function argument
"developer training?" Are you kidding?

That being said, Andre, there's a basic reason for doing this. If you use the same code pattern over and over again, and you have to copy/paste a lot of references, it's much easier to maintain code that's been structured this way. It's ever-so-slightly harder to DEBUG it, but it's substantially easier to EDIT it.

This, of course, wouldn't be necessary if @functions had a serious editing tool. Perhaps DDE will finally give us the opportunity for that.]]>
"developer training?" Are you kidding?

That being said, Andre, there's a basic reason for doing this. If you use the same code pattern over and over again, and you have to copy/paste a lot of references, it's much easier to maintain code that's been structured this way. It's ever-so-slightly harder to DEBUG it, but it's substantially easier to EDIT it.

This, of course, wouldn't be necessary if @functions had a serious editing tool. Perhaps DDE will finally give us the opportunity for that.]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/macro-formulas-temporary-variables-for-each-function-argument?opendocument&comments#06252008055718PMWEBTRD.htm Wed, 25 Jun 2008 05:29:38 PM -0400 Charles Robinson Diatribe: Macro formulas, temp variables for each @Function argument
I'll be the first in line to set fire to Domino Designer, but the IDE being barely functional doesn't make the language worthless. Formula is very robust, very fast, and relatively simple. Yes you can get yourself in trouble with it in a hurry, and yes, the tools for working with it are atrocious. That's no reason to condemn the language, though.

As Andre has already said, there is a balance between pure performance and readability. If you use 6 variables set to static numbers to feed into an @Adjust it's a waste of time and CPU cycles. Just put the numbers into the formula already. However if you're doing something more complicated, or something you'll use repeatedly, a variable makes sense.]]>
I'll be the first in line to set fire to Domino Designer, but the IDE being barely functional doesn't make the language worthless. Formula is very robust, very fast, and relatively simple. Yes you can get yourself in trouble with it in a hurry, and yes, the tools for working with it are atrocious. That's no reason to condemn the language, though.

As Andre has already said, there is a balance between pure performance and readability. If you use 6 variables set to static numbers to feed into an @Adjust it's a waste of time and CPU cycles. Just put the numbers into the formula already. However if you're doing something more complicated, or something you'll use repeatedly, a variable makes sense.]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/macro-formulas-temporary-variables-for-each-function-argument?opendocument&comments#06252008052938PMWEBT83.htm Wed, 25 Jun 2008 04:20:09 PM -0400 Eric Romo Diatribe: Macro formulas, temp variables for each @Function argument
Those practices create confusing code structures that are further exaggerated by the non-standard (non-existent?) formatting, as well as inconsistent syntax highlighting and error-checking of the IDE.

Formula language is still not ready for prime time. That's fine, but then don't try to overcome those flaws, rather limit their downside by using best practices like keeping the routines simple, limiting the nesting of If clauses, and using meaningful variable names.]]>
Those practices create confusing code structures that are further exaggerated by the non-standard (non-existent?) formatting, as well as inconsistent syntax highlighting and error-checking of the IDE.

Formula language is still not ready for prime time. That's fine, but then don't try to overcome those flaws, rather limit their downside by using best practices like keeping the routines simple, limiting the nesting of If clauses, and using meaningful variable names.]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/macro-formulas-temporary-variables-for-each-function-argument?opendocument&comments#06252008042009PMWEBRUN.htm Wed, 25 Jun 2008 11:08:13 AM -0400 Craig Wiseman The mysterious disappearing blog entry
just noticed the URLs...]]>
just noticed the URLs...]]> http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/the-mysterious-disappearing-blog-entry?opendocument&comments#06252008110813AMWEBKRP.htm Wed, 25 Jun 2008 10:08:11 AM -0500 Andre Guirard Diatribe: Macro formulas, temp variables for each @Function argument http://www-10.lotus.com/ldd/bpmpblog.nsf/dx/macro-formulas-temporary-variables-for-each-function-argument?opendocument&comments#06252008100811AMAGUKQW.htm