Genii Weblog


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


Thu 24 Mar 2016, 02:03 PM
After the relatively simple data mining example I showed previously using the Midas LSX, I thought I'd show a more useful example. In this, I want to find duplicate file attachments, whether or not they have the same name. This could be tracking down how a particular attachment has been circulated, or simply finding out whether replies and replies-to-replies have duplicated the same attachment many times, wasting valuable space. This could obviously be done with any database, but I wanted to use a real example rather than a demo example.

The basic concept here is that I generate an MD5 hash value for each file attachment and save the hash value, the file size and the attachment name into a line in a CSV file. Again, no additional programming necessary, I just used the Export to CSV sample db. Here is what I found. The top red arrow points to the item I moused over. This is a file called licmidas.lic, an old license file for our Midas LSX, but clearly it has been sent a few times, or has been left attached as the replies and replies-to-replies have been used. 

Below that, the row of red arrows shows that the file attachments has different names. The numbers are the file sizes, all the same obviously for identical files. In this case, the same attachment has eleven different names. (These happen to be image attachments created by one of our customers every time they reply.

Inline JPEG image



By simply viewing the data, we can see the UNID of the documents where the attachments live, so we can go and deal with them.


Inline JPEG image


We can do the same with embedded images, as well as some other rich text constructs I will show another time. As an example of how the duplicate images could be useful, imagine that your company needs to track the use of a licensed image that appears on various web pages. This could ensure that they were all found, even though images don't have an internal name to make them easy to find.

As before, here is the Export Directive I created in the Export to CSV sample. (This uses a soon-to-be-released version of the Midas LSX.)

Inline JPEG image


Update: I created a more specifically useful visualization for my own email database, so I thought I'd add it here.

Inline JPEG image

Copyright © 2016 Genii Software Ltd.

Tue 22 Mar 2016, 09:41 PM
Every few years, our Midas LSX evolves to take on new challenges. Sometimes, those are challenges that seem obvious, such as exporting Notes data to HTML or JSON with high fidelity. Sometime, though, the challenges are not so obvious. Over the past year, we have been evolving the data export features to allow data mining, or selective extraction. While I could show lots of details, I thought I'd write a few posts showing different sorts of data extraction. All are done with the Export to CSV sample, and so require no "programming" even with LotusScript. For these, I'll generally use Tableau, an excellent data visualization tool, though honestly the output is a CSV file which could be analyzed in almost any software package.

For the first post, I thought I'd search for a specific phrase in rich text fields, although I will export more than just the phrase. Each row contains a text record along with the name and UNID of the documents, so if somebody mentions the phrase multiple times, there will be multiple rows. If there are no mentions, there are no rows created.

I did two simple charts, one for the phrase "rich text" and the other for the phrase "security". For both, I searched the Business Partner forum from 2004. (At the very bottom, you can see the Export Directive I used to get an idea how this works.)

Inline JPEG image


Inline JPEG image

The form I filled out for the "security" search is below.

Inline JPEG image


Copyright © 2016 Genii Software Ltd.

Mon 21 Mar 2016, 09:41 PM
It is exciting to see my friends in the IBM Partner community sporting their spiffy new badges, especially Theo and Gab with "lifetime achievement" badges.
 
But none for me. I'm not fishing for anything, or even sad. I'm not an ideal IBM partner. Sometimes, I show IBM up by making email look good when they cannot, or enabling functionality they have never seen fit to add. Sometimes, I am critical of their many shortcomings. Most of the time, I just don't play the game.
 
So, why bring it up? I guess because I am not alone among the many who make IBM successful and each of those badge-owners successful, but who don't get badges themselves. I've known a lot of IBMers and partners and ISVs and customers and developers and administrators, many of whom do their jobs and don't get noticed.
 
So, this badge is for you. You know who you are if you are reading this, though almost certainly you are not. The vast majority of you do not read these blogs or hang out on PlanetLotus. 
 
Nonetheless, this badge is for you. Have a great day!
 
 
 
 
 
 

Copyright © 2016 Genii Software Ltd.

Thu 3 Mar 2016, 12:38 PM
For the past 13 years, CoexLinks Fidelity has been very successful at helping companies with diverse email systems coexist. From handling doclinks in the early days to high fidelity rendering in recent years, CoexLinks Fidelity has been used by many of the largest companies in the world, as well as some of the fairly small.

Over the years, we have expanded the core functionality, but have always focused on email as it moves through the Domino system and out, whether "out" means outbound email to clients, customers, family and friends, or "out" means to mobile devices (via Traveler) and web access of internal email (via iNotes).

While many of our customers remain dedicated to Notes/Domino email, many others are looking at IBM Verse, Microsoft Exchange, Google Gmail or other offerings in the diverse marketplace today. To continue to provide for those customers, we are expanding the reach of CoexLinks Fidelity to process email in batch mode. Of course, CoexLinks Fidelity will continue to handle all of the email rendering as it does not, including the very popular Message Store feature that customers depend on to isolate and make accessible content which is encrypted or otherwise needs special handling.

What we are adding falls into two categories:

1) Conversion in-place. An administrator can simply target a user's email database and convert some or all of the emails which are in rich text to high fidelity MIME such as we produce now with CoexLinks Fidelity outbound mail. The primary purpose is to allow migration of email to IBM Verse. Other migrations will handle putting the email in Verse but with the same deficient rendering that the Domino server has now (worse even than the Notes client rendering). With CoexLinks Fidelity, that email can be pre-converted to retain its Notes fidelity in Verse.

2) Export of email. An administrator can likewise target a user's email database and export some or all of the emails, including those already in MIME, to either a MIME stream or disk files. This is primarily used for migration to other email systems and for archival purposes outside of the Domino infrastructure for long term accessible email retention.

CoexLinks Fidelity 4.00 will surface some of this functionality next week, with additional options available over the next few months. If you have a use case you would like to discuss or want more information, please give us a call at +1 216-991-5220 or send us an email at sales@geniisoft.com. As always, current customers who are up-to-date on maintenance will receive a free upgrade. If you are not a current customer, but would like to try out an evaluation, fill out an evaluation request.

Inline JPEG image

Copyright © 2016 Genii Software Ltd.