Louis Kessler’s Behold Blog

Everything has its time. And the time for GenSoftReviews is up.

GenSoftReviews is a site that was at www.gensoftreviews.com that I ran from 2008 until today, to allow users to come and review their genealogy software. They could rate the software from 1 to 5 stars and leave a textual review with a suggestion that they also list what they thought the “biggest pro” and “biggest con” was.

The site would show you the programs most recently reviewed and their average ratings, ratings this year, last year and earlier, and allow you to view the individual reviews and search for a type of program or a specific program.

As a software developer myself, I was interested in following the offerings of other developers to see what people liked and disliked about them.

But over time, I’ve learned all I needed about the software out there. And interest in the site has started to wane, down from the nearly 1000 reviews a year from 2015 to 2018 to under 250 reviews a year in 2022 and only 114 so far in 2023.

I think part of the reason for the decline is that most genealogists have already found the software they are happiest with. Most programs have reached a point where they are as full featured as they can be and changes over the past decade to most programs have been minimal.

The same is true for computers. People are holding on to them longer and longer, simply because they continue to work well and there is nothing special in new computers that jumps out and says you have to buy me.

Shut Down

So today I shut down the GenSoftReviews site. This is how the home page looked at shutdown:

The entire site now redirects to a page I put on my Behold site. It gives a brief history of the GenSoftReviews site and suggests that users who are looking for a program to try one of those that have won a Users Choice Award. The final landing spot is now here:  https://www.beholdgenealogy.com/gensoftreviews.php

My Blog Posts About GenSoftReviews

It’s probably worthwhile listing here all the blog posts I’ve done where I’ve talked about the site and genealogy software in general. So here they are:

• GenSoftReviews - 24 Sep 2008

• GenSoftReviews User Choice Awards - 2 Jan 2010

• GenSoftReviews Big Update - 2 Jan 2011

• What Makes You Like A Genealogy Program?  - 11 Jan 2011

• Aren’t Today’s Genealogy Programs Any Better? - 27 Jun 2011

• Can Genealogy Software Be Rated Fairly? - 6 Jan 2013

• How Good are GenSoftReviews Ratings? - 4 Jan 2014

• Almost 800 Genealogy Programs. You’ve Tried How Many? - 27 Apr 2014

• Why are there so many genealogy programs? - 29 Jul 2015

• 1000 Genealogy Programs on GenSoftReviews - 18 Feb 2018

• Some Rogue Program on GenSoftReviews - 1 Apr 2018

• GenSoftReviews Users Choice Awards 2019 - 2 Jan 2020

• 2020 GenSoftReviews Users Choice Awards - 1 Jan 2021

• GenSoftReviews Award Winners 2022 - 1 Jan 2023

My Favorite Innovation

When I created GenSoftReviews, I customized much of the interface the way I wanted to be.

My most favorite innovation was the Exponential Weighting Algorithm I used to average the users’ star ratings. It would give every review a weighting based on how old the review was. A review one year older than a newer review would get 1/2 the weight of the newer one. A review two years old would get 1/4 the weight. This ensured that the most recent reviews would always have the greatest influence.

And I managed to do this in a single SQL statement that calculated the weighting on the fly each time an average rating was displayed. The SQL was:

SELECT
COUNT(comment_date) as num,
SUM(comment_date >= ‘$latestyear’) as num1,
SUM((comment_date >= ‘$previousyear’)*(comment_date < ‘$latestyear’)) as num2,
SUM(comment_date < ‘$previousyear’) as num3,
SUM(comment_rating*POWER(0.5,DATEDIFF(now(),comment_date)/365)) / SUM((comment_rating > 0) * POWER(0.5,DATEDIFF(now(),comment_date)/365)) as rating0,
SUM((comment_date >= ‘$latestyear’)*comment_rating*POWER(0.5,DATEDIFF(now(),comment_date)/365)) / SUM((comment_rating > 0) * (comment_date >= ‘$latestyear’)*POWER(0.5,DATEDIFF(now(),comment_date)/365)) as rating1,
SUM(((comment_date >= ‘$previousyear’)*(comment_date < ‘$latestyear’))*comment_rating*POWER(0.5,DATEDIFF(now(),comment_date)/365)) / SUM((comment_rating > 0) * ((comment_date >= ‘$previousyear’)*(comment_date < ‘$latestyear’))*POWER(0.5,DATEDIFF(now(),comment_date)/365)) as rating2,
SUM((comment_date < ‘$previousyear’)*comment_rating*POWER(0.5,DATEDIFF(now(),comment_date)/365)) / SUM((comment_rating > 0) * (comment_date < ‘$previousyear’)*POWER(0.5,DATEDIFF(now(),comment_date)/365)) as rating3
FROM $wpdb->comments
WHERE comment_post_ID = $post_id;

would produce the data needed to display this:

See if you can figure out how it works. 

Part of the work of genealogists in this digital age is to scan their old physical material. I’ve got many boxes in the closet and basement as well as binders in my shelves that still need to be gone through and digitized.

You can only do this one piece at a time, so it’s a manner of picking a project and working through it. Then going to the next.

4 Binders of Emails

One of the boxes I went through had a few dozen loose emails from genealogical correspondence that I had printed and meant to file one day. I was actually surprised to find these, because I had 4 binders of printed emails in my bookshelf, and I thought I had already filed them all into these binders.

Since December 2002, I have retained all my emails related to my personal genealogical research in a “Correspondence Genealogy” folder on my Windows computer in Outlook. Between then and now, I have been able to transfer my emails from computer to computer as I upgraded. In total I currently have 9,786 emails in that folder which works out to close to 500 a year.

But from 1995 to 2002, what I used to do was print out my research-related emails and I filed them into 4 binders.

Even if it was a larger photo, it would be hard to read what the faded labels say, but they say:

• G1A1a – My Genealogy Correspondence A-K
• G1B1b – My Genealogy Correspondence L-Z
• G1B2 – My Genealogy Correspondence (Unconnected) 1 of 2
• G1b2 – My Genealogy Correspondence (Unconnected) 2 of 2

The G1 numbering is part of my own genealogical file numbering system. I didn’t even notice the inconsistent numbers I had on these 4 binders until I wrote them above for this article. They should have been G1B1a, G1B1b, G1B2a and G1B2b.

Each binder contains my research correspondence with about 100 people I denote the start of each person with a tab divider that is simply one of those post-it-notes that come in pads. I find they work very well and you can get them in various colors. I don’t use the colors to denote anything. I just like the variation as opposed to them all being yellow.

Each tab lists the name of the person I corresponded with and below that is the family surname(s) and/or place(s) that the conversation was relevant to. Any person had between 1 and 30 printed emails during 1995 to 2002. So this would total another 2,000 emails or so that I received/sent during those years.

About half of those (the first two binders) are from people that are in my tree and I know how I’m connected to, and the last two binders are people that are researching the same surnames or places as me, but who as of yet are unconnected.

My Digitization Equipment

I thought it worthwhile to document the process I use. Someone may find it interesting or useful.

First thing of course, is the equipment. Since these are all printed emails on standard 8.5 x 11 inch paper, the best tool is a sheet feed scanner.

My current scanner is the Epson ES-580W. It’s about $500 and scans at 35 pages per minute.

It fits nicely on my desk and folds up nicely to be out-of-the-way when I’m not using it.

My previous scanner was a higher model Epson DS-860 costing $800 that scanned at 65 pages per minute, but as you’ll see below, it’s not the scanning speed that slows the work. It worked fine for about 5 years. A few years ago, it started occasionally adding lines to my scans, so I had to replace it.

The scanner comes with Epson ScanSmart software which works well enough for me.

The most important thing to me is that the software can save to Searchable PDF format. That way, I can use Windows to search all my saved PDF scans to find documents with specific words in them.

My Digitization Process

These are the steps I take:

1. Select all the emails with one person denoted by a post-it-note and take them out of the binder.
2. They were sorted by date with most recent on top, so I reverse the order so that the earliest is first.
3. Place the first email from the group in the scanner sheet feed.
4. On the scanner, select 1-Sided or 2-Sided (don’t have to select if it is the same as the previous scan) and press “Start”
   
5. The ScanSmart software opens and I can review that the page scanned okay. I then press “Next”.
6. Up comes a Window that says “Select Action” and I press “Save”.
7. Up comes a window that says “Save to Computer Settings” which I have set to “Searchable PDF” and I press “Save”.
8. I repeat steps 1 to 7 for each email with this person.
9. The searchable PDF files were all saved to my Documents folder as “scan_nnnn.pdf”.
10. I have my Documents folder sorted by latest date, so the scanned files will appear at the top:
    
11. I have two folders for these emails named “Correspondence Gen Connected” and “Correspondence Gen Unconnected”. In the appropriate folder, I click on the “New folder” button to create a new folder.
12. I rename the folder with the person’s name and in parenthesis the surnames and/or places that the emails pertain to. This will be the same information that was on the sticky note. e.g. “Clyde Hertzman (Hertzan)”. I sort of like putting the person’s given name first, but I assume most people will put the surname first.
13. I open up the person’s folder and move the 4 scanned emails from the Documents folder window into this folder.
14. I now rename each of the files with the date in yyyy-mm-dd format followed by the “type” followed by the subject line of the email. The type can be “email” if the email was from the person to me, “reply” if the email was from me to the person, or “message” if the email was from the person to a genealogy mail list or forum. e.g.:
    
15. I drop the original printed emails and post-it-note for this person into my recycling bin that’s next to my desk.
16. Now I repeat steps 1 to 15 for the next person.

This works well enough for me. Each person will take between 1 and 3 minutes to do depending on the number of emails.

Of course this will be slowed down further since I’m re-reading these 25 year old emails as I go and looking to see if there is relevant information that didn’t connect to my tree back then, but does connect to my tree now, 25 years later.

I find with this rechecking, that every so often, I get to add new information to my tree for people that I did not have in my tree back then but do now. In addition, several of these “unconnected people” back then are connected now and I’m sending out an email (when I can find them, since the email is rarely the same after 25 years) to let them know of the new connection.

These rabbit holes add time to the scanning process, but are so much fun and make the whole process worthwhile. Just scanning for scanning’s sake doesn’t cut it.

Summary

On average, in 1 to 3 hours each day I’ve been able to go through about 15 people. One binder has taken me a week. I figure I’ve got another 3 weeks to go to do the other three binders.

I’ll end up with two folders on my computer, one for connected people and one with unconnected people, containing about 400 people subfolders with about 2,000 emails in them.

All the emails are searchable PDF files. Here’s what a Windows search in File Explorer looks like:

In three weeks when I’m done with this project, I’ll pick my next one. And I’ll have 4 less binders on my bookshelf.

It’s been a long time since I blogged about the tools I use on my computer to deal with my photos. Looking back, I had these articles:

• Jan 2006: Windows 98 to Windows XP
  On Windows 98, I was using JASC Paint Shop Photo Album 5.  When I got a new computer with Windows XP, JASC had been bought out by Corel. I found that JASC Photo Album continued to work on XP without problem, so I continued to use it.
• April 2009::Picture Management and Editing in Windows Vista
  where I talked about my transfer from Windows XP to Vista. On XP, I had tried Adobe PhotoShop Elements, Google Picasa and Corel Photo Album. On Vista I was expecting to try Windows Live Photo Gallery which included face recognition.

At that point, I did start using Windows Live Photo Gallery and I really liked it.

The last release of Windows Live Photo Gallery was in 2012, and Microsoft stopped supporting it and made it unavailable in 2017. I continued using it on my next computer I purchased in 2014 with Windows 8.1.

• Mar 2020: When Everything Fails at Once
  My computer failed spectacularly, but fortunately the drive with all my photos and videos was fine. For it’s replacement, the Windows 10 computer I am still using today, I decided not to install Windows Live Photo Gallery, but look for something new, thinking that maybe Microsoft Photos which was included with Windows might be okay even if it did not have face recognition at the time.

Tag That Photo

Microsoft Photos just didn’t do it for me. It didn’t work the way I worked. I needed something better that I liked.

Then I ran across a program with a special purpose. It would import the thousands of tags that I added to my pictures in Windows Live Photo Gallery and write them as metadata in Adobe’s XMP format into each picture. This would preserver the thousands of faces and other tags that I had added over the years in photo gallery.

The program is called TagThatPhoto. It also has Face Recognition. After I got it and imported all my Photo Gallery tags, I then spent probably a month going through all my photos with the program’s help, to tag as many more photos as I could. My current count is sitting at over 18,000 faces tagged with another 24,000 being faces of people who are not relevant to me.

However, face tagging was really all TagThatPhoto was good for. It did not really help me at organizing my photos like Windows Live Photo Gallery did. So I was in need of something else.

digiKam

I wanted something that would do all the organization for me. I searched for several months for a program with face recognition that would allow me organize all my files into topics, include the ability to locate my photos on a map, along with tagging, captioning, editing and even a capability to find duplicates. There weren’t too many programs that impressed me.

Then I ran into digKam. Not only did it seem to have everything I needed, but it was Open Source and free as well. I downloaded it and put it through its paces.

You could see the tabs at the left are: Albums, Tags, Labels, Dates, Timeline, Search, Similarity, Map, People – basically all I needed.

But it turns out digiKam for me isn’t the right program. I just don’t like the user interface. I found it slow in rendering the photos and difficult to get around. It really is a Unix program that is ported to work also on Windows and to me it felt like it.

Mylio

So then it came to be time for RootsTech 2023 in Salt Lake City. I’m not attending this year but was looking very forward to its live streamed and pre-recorded sessions.

What I discovered was one of the conference’s sponsors: Mylio Photos. The more I watched their explanatory videos, the more intrigued I became. Here is a full-featured photo organization program that I was very surprised I had never heard of before. Checking out the Mylio website (which is beautiful by the way), I was surprised to see they are up to Version 22.1. So they are not a new company and have been around for a while.

I am very impressed by the program’s advertised capabilities, the user interface of the program, and the company’s videos and presentation of their program.

This is not a free program. They charge $99 a year. But if the capabilities are this good and if I enjoy using it as much as I did Windows Live Photo Gallery, then it will be worth every penny.

Tonight, I will install and try it out.

Evening progress:. I bit the bullet. Purchased a year subscription on their RootsTech special offer ($20 off + 2 bonuses), installed it and started loading my photos.

I’ve got 2 main folders with photos, one in my Pictures directory and another that is mostly trip photos that are in my OneDrive folder on my computer that get synched with OneDrive. My Pictures directory had 30,198 pictures and videos in it. They took about an hour for Mylio to index and then started to generate thumbnails, detect faces and scan for text.

I think this program’s a keeper.It does so much. Rather than me try to explain, I’ll leave you with this RootsTech presentation by Rich Harrington from Mylio.

1. It does not write the XMP into the image file.
2. It does create an XMP “sidecar” file with the Mylio information in it.

Not having the XMP info include the face regions and other changes in the file is not a good thing. If the picture is moved without the same-named .xmp file, then the info is lost. And a lot of programs can’t read the sidecar information.

By comparison, digiKam directly writes the face regions as XMP information into the file and makes changes directly to the file as well. Mylio can read this information, which is good. But digiKam does not read Mylio’s sidecar information. It must be a proprietary format.

I do understand the possible logic that Mylio used in not adding the info directly to the file. With the sidecar file, they do not change the file in any way, and all edits and additions can easily be undone – and they state that.

But if portability is needed and you want those faces and edits to be available to any future photo management program you may decide to use, then Mylio might not be the best. It’s a bit of vendor lock-in which isn’t good.

And having thousands of little .xmp files all over my computer is something I don’t want. I’ve also noticed some 256 x 256 pixel copies of pictures appearing in the same directories.

Those look like thumbnail copies of the full size pictures that Mylio might be using. If a program makes thumbnails, it should keep them in a database that doesn’t add files to someone’s pictures folder.

So I’m afraid it might be back to digiKam for me.