Ben Langhinrichs

Photograph of Ben Langhinrichs

E-mail address - Ben Langhinrichs







Recent posts

Mon 16 Sep 2019

About that email in Notes



Mon 9 Sep 2019

Perils of PDF 4: Missing and obscured data



Fri 6 Sep 2019

Perils of PDF 3: Wide Tables and data loss


September, 2019
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

Search the weblog





























Genii Weblog

HCL is doing so many things right, but...

Wed 3 Jul 2019, 11:01 AM



by Ben Langhinrichs
Overall, HCL has been doing a terrific job. They have revitalized Notes/Domino and fixed some longstanding bugs. They have worked with business partners better than IBM has in a long time. They have injected hope and enthusiasm, and have mostly met all the deadlines they've set for themselves. They have shown vision and commitment to future versions, and I see customers responding.
 
But still I worry. If there is one lesson that large companies who create "platform" products should have learned by now, it is that an ecosystem of third party products, addons, and integrations is key to survival. No one company, no matter how big or dedicated or enthusiastic, can do it alone. There need to be ISVs to meet the niche needs, to innovate in areas HCL can't afford to focus on, to take risks that HCL can't.
 
 
Inline JPEG image
 
 
Lotus understood this, and for the most part IBM did as well. One of the great reasons why Lotus Notes became a huge success early on was a surprisingly robust API that was maintained and enhanced with each version. At times, there were even multiple APIs including the venerable C API and the easier-to-learn but less complete C++ API.
 
The most recent Notes C API was released in 2013, coinciding with the Notes 9.0.1 release. Five and a half years ago.
 
Notes/Domino 10 was released in November 2018, seven months ago. There is still no updated Notes C API, nor any promise of when one will appear.
 
This was all less of an issue for the years following 9.0.1 because frankly, the old toolkit worked fine and nothing was changing. But things are changing now, including new features such as DQL as an example. ISVs watching from the sidelines see features they could build on... but can't. Integrators who might be newly enthusiastic about integrating with Salesforce, Microsoft Flow, whatever are going to hesitate when the most recent deep integration is so far out of date. Along with the C API itself, the related toolkits were built for compilers now considered ancient. To make matters worse, not only is there not a new API or new toolkits, but the commitment to the always-up-to-date API has come into question. If there is no comprehensive API available for Notes/Domino 10, which is at least mostly like Notes/Domino 9.0.1 under the covers, will there be one for Notes/Domino 11, which is likely to have some more substantive changes?
 
Of course, there are many products which require only LotusScript or features available in the product itself. HCL has made some promising noises about app stores and product delivery for those sorts of products, and that is great. But those tend to be products which live entirely within the Notes/Domino platform, and the world is far too integrated and complex to be limited to those. Tight integrations at various levels with non-HCL products tend to require lower level APIs.
 
So, while I know resources are stretched and everybody at HCL is working incredibly hard to deliver on the promises made, I also know that HCL risks a lot by not following through on the longstanding promise of the Notes C API. While I know that very, very few developers use the Notes C API directly, it is a key to the ecosystem of extensions and products which business customers need to actually embrace Notes/Domino again.

Copyright © 2019 Genii Software Ltd.

What has been said:


1098.1. John Curtis
(07/03/2019 11:15 AM)

I have forwarded this internally. It has forever and always been a struggle to commit to which API calls to effectively "freeze" by publishing. It's not just an analysis of what exists, has never changed, etc., but a calculation of support cost.

I can speak to the DQL calls - we're just not done to a point that allows it yet. We will be.

Another forever and always thing is the estimation by people of the capacity of the team to provide what is asked for. I mean simple horsepower capacity. Without going too deep, I will just ask a telling question - 2 year ago, do you you IBM added to our number prior to the first HCL transaction?

Don't hear me making excuses, just stating fact.

We'll get there, and consider a fire under this particular issue. Thanks so much for your commentary!

-John


1098.2. Ben Langhinrichs
(07/03/2019 11:28 AM)

Thanks, John. I appreciate your forthrightness and commitment to getting there. I also understand that the team is handling a ton of stuff already, and that the C API is always a moving target when new things are being built as it is hard to know what to put in and what not to. But I also know that the squeaky wheels tend to squeak about more visible issues, and want to make sure it doesn't get forgotten.


1098.3. Hogne B. Pettersen
(03.07.2019 12:13)

Excellent writeup, Ben. I fully agree with you. The number of third party products have been slowly decreasing since 2012. Most of the ones I was administering are dead and on life support in the companies using them. Too bad.


1098.4. Ben Langhinrichs
(07/03/2019 12:34 PM)

Hogne, I have high hopes that if HCL continues to push and enhance the product and win over big customers, some of the older third party vendors will try again, and maybe even a few new ones will give it a try. Granted, many got burned by the neglect over the past few years, but the opportunity to make money heals a lot of old grudges, and HCL is not IBM.


1098.5. John Detterline
(07/03/2019 02:42 PM)

I agree with this post. A former client that was a long time Notes/Domino client, used the lack of integration with Unified Communications products as one of the reasons to dump the entire platform. It is very difficult to justify using a product when you can't point to other products that work with it. Thanks for writing this.


1098.6. John Head
(07/03/2019 03:46 PM)

@John Curtis - it's far better to get a new version out for each major release - and then change things later for the next release, than to skip releases for the API Toolkit. Features are never done. We all understand that HCL is working their asses off and catching up from the lull in the past few years is difficult. But there needs to be an API Toolkit release EVERY NOTES RELEASE, period. Not doing it is detrimental to the adoption of ISV's.

Have your say:

Name *:
E-mail:
e-mail addresses will not be displayed on this site
Comment *:


<HTML is not allowed>
Linking: Add links as {{http://xxx|title}}, and they will be activated once approved
Blocked? Unable to post a comment? Please read this for a possible explanation...