Get the most from Notes®

1 - I answered more at Ed's blog, but ask IBM about their dbs
Some long time companies, banks/insurance/pharmaceuticals usually have tons of dbs.
It can be contained and dealt with relatively easily if you have the tools and know what you are doing.
It could be worse when I was with Bankers Trust in the R2/3 days we had 100's of servers, then an equal number of backups!
I can imagine they fall into the higher end of the numebrs.

2 - I'd be interested in the KPI of volume of email per user in relation to the Notes Application per user would be.

In some organizations, the use of Notes Apps ELIMINATED the need to send so many email messages, which can be a very good thing. Discussions go in the discussion DB, or team room, etc. Docs go into the DocLib, Workflow apps may have many task queues, etc.

I don't see how one could jump to a conclusion that the high-ratio of NSF/User is good or bad, without knowing how that KPI fits into other KPI's within that entity, their workflows and business processes, or thier industry.

3 - OBTW, my org has one user, a handful of Domino severs running at any one point, and hundreds of NSF-based apps, but that reflects the special needs and circumstances of my work.

-------------
Over the years I have observed that organizations that "get-Notes" are more likely to have invested in the platform and less likey to think of Notes as "just-email". The correlary is that organizations that are "slipping away" are email-centric, and would likely not have very many NSF-based apps.

From that perspective, the KPI for NSF'/User could be an indicator of the degree of "switching difficulty".

From a SysAdmin/AppDev perspective, it would all depend upon how well you managaged the situation, and if the apps are well-designed, or if you have some other business reason for not wanting so many (or so few) NSF's.

There may be valid ACL-Security reasons for having a large number of NSF's.

I'm working on a design that may have 300+million NSF's, but fairly few people. These NSF's will have the same basic design, but for security (and performance) reasons, this partitioning is the best solution approach.

4 - Bill this sounds like a proposal Lotus once turned down for a European Country which wanted to give every person their own email mail box.
Or in your case, perhaps yoou are trying to handle email reporting in a more efficient manner. Sounds like a very cool project, would love to hear about it.

5 - Hi Keith,

My proposition is a unique public-private solution that is community-based and nationally-scoped. The subject matter is Life-time Personal Health Record System, which is the holy-grail of the healthcare system. Email, PIM and productivity tooling is a byproduct of using the ND8 platform, and serves as a basis for ISP/ASP for-profit business models. There is a dual-paired non-profit business model to address the social-acceptance issues with the PHRS application rollout. We developed the initial prototype 15 years ago, but was a solution ahead of its time. Today, the industry is ready, and the ND/NSF archictecutre is ideally suited for this application.

Sadly, my attempts to get traction within the IBM/Lotus blogging community have come to nothing. It is a big concept with enourmous potential, and has the potential to swiftly change the tide w/r/t to penetration.

If IBM/Lotus doesn't get it, some of the investors think we'll just grab back the Lotus Division from IBM and do the right thing. Half Joking on this.

The real issue, I believe, is that IBM' IGS Consulting division is too committed to profit-making solution approaches that are driven by data center / database centralization, rather than being driven by consumer-centric data consolidation and data center decentralization, yet retaining a national research functionality capability.

Contact me directly if you want to know more.

6 - As Bill says, collaborative applications can, and sometimes must, be divided to achieve performance. I've seen applications, used by 20.000 users, organized as a small workflow of 30 databases.

Following this, if the company uses 500 applications with this distribution, it would have 15.000 databases.

Not always big ratio indicates bad administration.

However, I would recommend having an auditing tool to know if an application is being used and by whom.

7 - I would like to clarify one point from the original post. In the examples I cited (100,000 users / 50,000 dbs and 4,000 users / 8,000 dbs), the numbers excluded all templates, all mail files, all dbs on development or test boxes and all local dbs. So, these figures represent real apps on production servers.
Of course of the remaining applications many will be the same. Some organisations have hundreds of teamrooms all based on the same template (except that these dbs are often customised in the nsf . . .)
I wonder how many people out there feel confident that even half of their apps are still being used.

BTW, this point is also being discussed on Ed Brill's site:
{ Link }

8 - Bill, This blog system is not taking the emails or web sites so please email me, keith at keithbrooks.com.