Ben Langhinrichs

Photograph of Ben Langhinrichs

E-mail address - Ben Langhinrichs







Recent posts

Wed 7 Oct 2020

Custom Domino Extensions in a Modern AppDev World



Tue 22 Sep 2020

Exciton Boost - Barbarians are at the gate



Thu 17 Sep 2020

Exciton Boost - WYSI(hopefully)WYG


October, 2020
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

Progress?

Fri 5 Oct 2007, 05:57 PM



by Ben Langhinrichs
I know I am a bit of a nut about HTML rendering, but look at the image below:

Table rendered in Notes 8Table rendered in Notes 8

Here is the same table rendered by our Midas Rich Text LSX 4.00

Rendered by Midas Rich Text LSX

They look pretty similar, right?  I first created this sample was back in Notes 6.x days.  I don't even have a Notes 6.x version on this machine, but I do have Domino 6.5.3.  Here is how that same table renders in Domino 6.5.3 (and how it rendered in Domino 6.0.0, for that matter):

Rendered by Domino 6.5.3
So here is the kicker.  Here is how the same table renders in Domino 8.0.0:

Table rendered in Domino 8

Um, progress, right?

Copyright 2007 Genii Software Ltd.

What has been said:


631.1. Carl Tyler
(10/05/2007 07:22 PM)

Notes/Domino 8.0.0 was a client based development effort blah blah. That doesn't really matter much for those companies that swallowed the Portal story of a few years ago and saw the browser as the new client and have been converting all their Notes apps to Domino web apps.

Domino Web Server is really starting to show it's age. The loops we as developers have to jump through to make the notes client and Web Server have some parity is quite frankly ridiculous, especially now that HTML, CSS, XML, XSL support in browsers has improved so much. Sure there are some things rendered in the Notes client that don't make sense rendered the same on the web, but for most stuff it should just appear the same, why today, 10 plus years after Domino or the HTTP stack used with notes prior to Domino, do we still have to manually create simple things like date pickers ourselves? Why if I set a notes form to have a date field with the date picker doesn't Domino do the right thing for the web?


631.2. Ian Randall
(10/06/2007 04:32 AM)

Hear, Hear. I totally agree.

My company has struggled for many years to maintain full functional parity between the Notes User interface and the Web browser interface to our product.

When Lotus developed the INotes interface into the mail template (arguably a pioneering Web 2.0 initiative), I hoped that the significant development effort that Lotus undertook to make their own product match the functional parity between the rich client and the web browser for email would flow-on to the Notes IDE and spur them to improved HTTP rendering engine.

But no that didn't happen. Even now with all the great things that have been achieved with the Notes 8 user interface, there appears to be little progress made on the Web browser interface into Domino.

Personally, I would be happy if Lotus simply gave us full functional parity between the Notes Client and the Web browser, even if the UI remained a little ugly.

I also feel they would be better off making the Domino Designer Client open source and allow the Notes Development community the opportunity to fix it ourselves.

Ian


631.3. Bruce
(10/06/2007 08:47 AM)

Ben - can you please fix the link in the above comment? TIA


631.4. Richard Schwartz
(10/06/2007 12:52 PM)

Call me dense, but the captions on pictures 1 and 4 seem to be saying the same thing, yet the pictures are different. Clarify?


631.5. Ben Langhinrichs
(10/06/2007 07:53 PM)

Rich,

Sorry, I didn't use the computed title for that version, so they do both say the same. I have modified the images slightly to clarify.

Ben


631.6. Richard Schwartz
(10/07/2007 06:36 AM)

Ah... that makes much more sense.

-dense


631.7. Ben Langhinrichs
(10/07/2007 07:15 AM)

Bruce - I have read Alan's comments, and while I have great respect for Alan and high hopes for the future, it doesn't stop the fact that it was obvious before Notes 6 was released that the HTML rendering was badly flawed and made Notes/Web parity very difficult. Waiting until 8.next is fairly inexcusable, imo, although it has made a good business for me with Midas. - Ben


631.8. Nathan T. Freeman
(10/09/2007 10:51 AM)

I'm pretty sure that Balaban has made it clear that changing this stuff is a target for NMFR.

Ben, if this is a priority for you, why haven't you communicated that to the team that can do something about it? You ARE a Design Partner, after all.


631.9. Ben Langhinrichs
(10/09/2007 06:55 PM)

Nathan -

I'm not sure I catch your drift. What makes you think I have not commented on this extensively to many of the people who could make a difference? Also, do you really think that being a Design Partner is going to stop me making comments like that here? I discovered this when I was tweaking a Midas setting and I went to Domino to see how it handled the issue (basically collapsed vs. separated models working properly in Firefox but not in IE). I was appalled to see that Domino doesn't handle any of it, so I posted this. Even if IBM were to fix it in NMFR, which I know Bob Balaban would love to do, I have little confidence in the completeness of their solution. I hope I am wrong, and I hope I can have an influence. Sometimes in the forum, sometimes here.

- Ben


631.10. Nathan T. Freeman
(10/10/2007 09:36 AM)

I guess I just don't see what your point is in observing that an already shipped version doesn't have the feature set that you want, when no one ever claimed it would, and you know perfectly well that it's planned for the next major release.

I guess it just seems like blogging "y'know, I was on my way to work today and I realized that my car isn't a Ferrari. MAN! CAN YOU BELIEVE THAT!??! I should have a Ferrari by now!"

It's your blog -- post what you want. As you say in your claimer (it's not a DISclaimer, is it?) I'm just confused about how you might think this would influence the outcome of NMFR, since it's already in the plan and you know that it's already in the plan.

Please understand -- I'm not trying to suppress discussion. I'm simply confused about what your after.

For the record, before I responded I specifically went out to the forum we both share where the guy that's writing the HTML rendering stuff for NMFR participates on a daily basis.


631.11. Ben Langhinrichs
(10/10/2007 10:59 AM)

Nathan -

I'm going to try to say this simply. I SELL A PRODUCT THAT FIXES THE PROBLEM. OK, that may be a tough concept, but there are a lot of customers who who will be using some version of Notes/Domino prior to NMFR. In fact, every single Notes/Domino customer uses a version prior to NMFR, because it doesn't exist yet. So, why would I bother posting this? Because I am a greedy capitalist, and I hope that people will look at the first image, then the second, and see that MIDAS DOES A BETTER JOB THAN DOMINO RENDERING HTML. Believe it or not, that is a bigger concern for me than whether or not NMFR fixes this issue or not. I fix it now, and in fact fixed it about seven years ago, so I have customers all over the place taking advantage of our far better HTML rendering. I run a business. We make money from customers who want better rendering now, not in some potential change in some potential future, but NOW. And in version 6.0.3 or 6.5.2 or 7.0.2 or 8.0.0 or whatever they happen to be using. NOW, not in the future.

So, what am I after? Customers! What do I gain from posting this? Business! What is my purpose? Exposing the weaknesses of Notes/Domino and the strengths of Midas so that people who use the former will also use the latter.

Is that simple enough?

- Ben


631.12. Ben Langhinrichs
(10/10/2007 11:29 AM)

P.S. I have personally been told by IBMers before every major feature release since R5 some variation on "HTML rendering will be better in R.x" and "wait until you see the great things we are doing with HTML rendering". I have read public comments to the same effect on several occasions. In addition, there is an implicit promise that IBM will keep up, and that promise has not been lived up to. I have high hopes that this NMFR will be different, and I will do what I can to facilitate that, but I'll bet you a chocolate milkshake at Lotusphere that when NMFR comes around, Midas will still have better rendering, and we will be told (as we have been about composite apps, for example), to "Just wait, the best is around the corner." and "We are still working on this" and "what do you expect in a first release". I sincerely believe IBM means well and will do its best, but the inertia of getting changes into the system makes agility very difficult.

In a related point, would YOU have predicted that rich text programmability in Notes 8 would have zero improvements? Even given that they were not promised, don't you think that is suprising? But why would you think that? Because there is an implied promise to "keep up", and that is what has been missed with HTML rendering. Not an explicit promise, but a pretty reasonable implied promise. And it is not just because Notes 8 is a server based release, as these fixes were not made for 6.0, 6.5, 7.0 or 8.0, even though the changes are actually fairly simple to make. Why do you think IBM should be let off the hook about breaking the implied promises? It hardly seems like you.


631.13. Nathan T. Freeman
(10/10/2007 11:41 AM)

:-)

So the "having an influence" is a smokescreen. You're really posting this to boost Midas.

:-)

That's cool.

And no, we don't have to bet the milkshake. Why would you think you have to convince ME that a BP or someone in the general community can do a particular thing better than Lotus does it!??! I'd already agree wholeheartedly with that.

However, I'd be happy to discuss this further at 'sphere over milkshakes.


631.14. Ben Langhinrichs
(10/10/2007 11:48 AM)

Nathan -

It is a company blog. The interests of the company are paramount. My personal feelings can be expressed elsewhere. :-) Incidentally, where did I say I was interested in "having an influence"? I have been pretty clear that what I want to do is promote Midas, CoexEdit, CoexLinks and OpenSesame. This post is just one of various that tweak IBM for various failings which our products, not coincidentally, happen to solve or improve. It's all about business.

- Ben

P.S. I'd be happy to discuss this, or anything else, at Lotusphere over milkshakes.


631.15. Nathan T. Freeman
(10/10/2007 12:10 PM)

"I hope I am wrong, and I hope I can have an influence. Sometimes in the forum, sometimes here."

645.10


631.16. Ben Langhinrichs
(10/10/2007 02:46 PM)

Ahh, no wonder the quote didn't sound exact. I do hope I can have an influence for the benefit of Notes/Domino, but that is because anything that strengthens Notes/Domino strengthens my customer base. Don't kid yourself, IBM is in it for the money as well, and they will do whatever helps their bottom line. I am no different, and will do what helps our bottom line, within the bounds of what I feel is ethical. Since pushing IBM to have better Notes/web integration makes a stronger market for my (superior) Notes/web integration tools, I'll certainly do what I can to be a positive influence. But it is not altruistic. It is purely mercenary.


631.17. Mika Heinonen
(10/12/2007 02:19 PM)

I learned from Nathan, that the reason why Domino Web Server does such inefficient output, is because that IBM believes that Netscape 3 must be still supported.

It would be however nice to have an option to say in Domino that "Generate code which works only in Firefox 2.x and Internet Explorer 7.x", it would be no problem for me to redirect the users to a website which sells new hardware which is needed to run Windows XP/Vista instead of Windows 3.1/Windows 95.


631.18. Nathan T. Freeman
(10/16/2007 08:48 AM)

That isn't what I said. What I said was that the HTML table code was written when Netscape 3 was a dominant platform, and that was one of the only options available. The reason it STILL does it the same way is because they haven't refactored that code.