Ben Langhinrichs

August, 2004
SMTWTFS
01 02 03 04 05 06 07
08 09 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31

Search the weblog





























Genii Weblog


Civility in critiquing the ideas of others is no vice. Rudeness in defending your own ideas is no virtue.


Tue 31 Aug 2004, 11:28 AM
I am working on an advanced version of our Report It! sample db, and am trying out a technique I tried shortly after Notes 6 came out but haven't touched in a while.  I basically use a bunch of layers to make graphical charts that work for both the web and Notes.  Here are three graphs in a row.  The first is an image showing the graph as it appears in Notes.  The second is an image showing the graph as it appears in IE 5.5 on my machine.  The third is the HTML generated by 6.0.2CF1, which you will see as whatever your web browser renders.  Can anybody tell me how it looks for them?  I will try other browsers and such eventually, but I'd love to hear any quick feedback.  (Even, "you can do that in Notes?").


Scatter graph in Notes (captured as image)


Scatter graph in IE 5.5 (captured as image)


Scatter graph in your browser

0
25
50
75
100
125
150
175
0
25
50
75
100
125
150
175






















Copyright © 2004 Genii Software Ltd.

Mon 30 Aug 2004, 07:31 AM
Reports about the large march in New York contain the same quaint estimates of crowd numbers that they did in the 1960's.  The marchers estimate there were four hundred thousand, the police estimate two hundred fifty thousand, and the New York Times estimates one hundred thousand.  The biggest change is that the police usually way undercount these marches.

What boggles the mind is why we have these wild estimates.  Technology should have rendered all of this obsolete.  With satellite images and high speed computers, it should be fairly easy to count fairly accurately.  I have seen news shows do this for nonpolitical events, so why do we retreat back to wild guesstimates when the event is political?

Copyright © 2004 Genii Software Ltd.

Wed 25 Aug 2004, 04:34 PM
My daughter, about whom I have written a few times before, is leaving for college shortly, but living with her has always been an adventure.  Some may remember her recent appearance on ABC (in which they showed her getting her prom picture taken, and my taking it as well, so I was on too, but more briefly):
My daughter, date and ABC news

Well, today, my daughter was working at an event hosting John Edwards for a town hall meeting in Cleveland.  Here she is with John Edwards (below):
My daughter with John Edwards

You know the amazing thing?  Everybody else at these events is there to see famous people, but my daughter is there to be seen.  First, Dennis Kucinich recognizes her and goes over to say, "Hey, you used to volunteer for my campaign."  Then, one of the Secret Service men came over to her and said, "I know who you are.  I saw you on TV last week".

See what I mean?  Stand back, world, because my little girl is coming at you.  Don't say I didn't warn you.

Copyright © 2004 Genii Software Ltd.

Tue 24 Aug 2004, 11:12 AM
There are a number of questions which come up on the forums over and over, and many of them relate to hide-when formulas.  I recently wrote a post on Writing better hide-when formulas, which has been very popular and is frequently visited and linked to.  I decided to follow that up with a post on hiding a rich text field, since that is also a common question.  I pulled this from a few of the many times I've answered this question, including this one back in 1999.

What is the question?
The question is asked different ways, but the common issues are that the hide-when formula appears not to work at all, or it only hides part of the rich text field.  Sometimes, attachments are shown when the rest of the rich text field isn't.

What is the issue causing the problem?
In my Rich Text 101 - Hide-when formulas and my earlier Rich Text 101 - Paragraphs, I explain that hide-when formulas are attached to or associated with paragraphs, and that since rich text fields contain embedded paragraphs, those can have hide-when formulas which differ from the form.  In particular, it is not uncommon to create a form with a rich text field and a hide-when formula on the field, then later change the hide-when formula.  On documents created prior to the change, the first paragraph is likely to reflect the new hide-when formula because it is the "default" paragraph, but subsequent paragraphs are likely to reflect the older formula.

What is the best solution?
Keeping in mind that there are no perfect solutions, there is one good workaround.  Put the rich text field inside  a section on the form, and add the hide formula to the section.  The best way is to collapse the section, and put the hide-when formula on it directly.  Now, expand the section and make sure there is no hide-when formula on the field inside the section.  This technique will hide attachments and both old and new documents, even if they were created before the change.  It also has the advantage that it does not add any hide-when formulas to the rich text inside, so the rich text can be copied elsewhere and not carry long unwanted hide-when formulas.

What are other solutions and why are they not good?
The most common other solution I see is to use computed subforms which contain the rich text field.  While this seems like a good idea, Notes has so many problems with rich text fields on subforms that you are likely to cause more problems than you solve.  Another solution people suggest is to use a computed field which computes to the real rich text field.  This just doesn't work, both because the computed formula will only work when the document is created, and because computed rich text fields can totally mess up the formatting of the rich text, stripping out tables and otherwise causing major changes you don't want.


Copyright © 2004 Genii Software Ltd.

Mon 23 Aug 2004, 11:28 PM
We have a new kitten, and in honor of the Olympics, my eight year old son named him Perseus, but we pretty much all call him Percy.  We have had a lot of cats over the years, but none for a couple of years, and not a kitten for many years before that.  Needless to say, things are not as quiet as they were.  I'll post a picture soon.

Copyright © 2004 Genii Software Ltd.

Fri 20 Aug 2004, 09:21 AM
I grew up in a family as one of four brothers with no sisters, so I missed out some on the peculiarities joys of dealing with women's issues.  Now, I have a daughter and a wife, and while they are both sensible and reasonable women with good heads on their shoulders, I have been exposed to more of the women's magazines, and what they tell women they ought to care about. One of the truly frightening intriguing concepts the magazines tout is the idea of "wedding weight".

As far as I understand the concept, women go to tremendous effort to diet and exercise before they get married, so they can "look good" for their wedding photos.  They frequently weigh less than they did in high school.  All that I can sort of understand, as I certainly wanted to look good for my wedding.  But then they weigh themselves, and immortalize that weight for the rest of their lives as their "wedding weight", and refer to who far off their wedding weight they are at any point.  After that, they can have kids, work jobs, grow to middle age, and generally live real lives, but none of it is supposed to weigh them down.  Guess what?  It does weigh you down.

Anyway, when a software product is new and building a following, nothing is more important to its developers than making it work the way people want.  Product launch is the first wedding, followed by a long series of mini-weddings as versions are added and the product is brought to perfection.  New customers love the product, and love the versatility and agility of the developers, who seem to read their minds and add new features just as they are needed, and fix bugs before they are too big a problem.  The better the developers do this, the more loyal the following.  A set of expectations is being built, both by the developers and the customers, which is the "wedding weight" of the product.

Then a few years go by.  If the product is successful, especially if it is very, very successful, a certain weight is added.  Any new feature must be tested against every other new feature.  There are now thousands, or even millions, of people using the software, and every fix seems to break something else.  API's change, standards change, and everything is supposed to keep working both with the old stuff and the new stuff.  Development times extend, reported bugs rise, customers wonder where the early responsiveness is.  Still, the added years have also worn down the rough spots.  The accumulated bug fixes have made a very stable product.  The features work together with a smooth, if sometimes tired, familiarity.  The customers are used to the "way things work".

But a few years later, more weight is added.  The original developers have all moved on to new challenges.  The new developers aren't as familiar with the "way things work" as their customers.  They have new ideas which may go against expectations.  Also, as with people, there are children.  Products are spun off which require nurturing, often at the expense of keeping the original "wedding weight".  Developer time is often split between the parent product and the children.

So, here we are with Notes/Domino (it used to be one name, but like a double chin, even the parent has expanded some).  Along with the weight of years of backward compatibility, adherence to once and future standards weighs on the hips like too many macaroni and cheese dinners.  The developers have mostly moved on, replaced by new developers.  The focus is split between the children and the parent.  And IBM's response is to roll out the rich client.

Now, the rich client is attractive, if not exactly slim.  It does not have the years of history, the weight of children, the drag and sag of endless customers.  But what is it really?  Is it a new and better Mom who, unburdened at last as the children have gone off to college, has started swimming and attending classes and looks happier and healthier than she has in years, ready to head into a new phase in life?  Or is it another woman, an attractive vixen who pretends to have the experience and wisdom of Mom, but who tempts developers with the skinny thighs of inexperience, the slender waist which comes when backwards compatibility is shed, the large assets of a new development environment?

Is the rich client a re-invented Notes, getting back to its wedding weight, or a pretender eager to inherit the wealth and reap the riches earned by its predecessor?  

I guess you'll have to answer for yourself while I brace myself for those who feel this whole article is sexist, which it really isn't meant to be.

Copyright © 2004 Genii Software Ltd.