Saturday, February 27, 2021

Rootstech

TL;DR -- We will pay attention to technology in the small and in the large, especially that related to advanced software. 

---

We love Family Search and have used it since we started this work which was over a decade ago. At that time, some of the interfaces for genealogical packages were still fairly kludgy, so we did our own thing with respect to collecting and storing data. Now that we have more curating work to do, some decisions need to be made. A basic issue? 

There is one choice that people make. They can either go whole hog into someone's care by using a heavily GUI'd system. Or, they can take a more independent approach, as we did. There is a third way which balances the two. Many packages, in their latest manifestation, seem to offer better support. But, we like what WikiTree represents a solution to form and interface. The data is easily accessible, however one also has access to the mark-up level. Granted, this is not code, but it gives one the feel. On the other hand, access to support app extensions will be important, too. 

This year, for the first time, we paid attention to Rootstech which offers the framework with which to look at some of these issues in terms of genealogy work. We will browse this information later; too, we will pay more attention, especially to discussions about the future. 

Another note about the future? How AI will play in all of this? What is called the 'deep fake' approach now can generate photos of people who never existed. Too, it can create videos of characters who are not real doing things that seem natural. Some of the current results are easily analyzed with respect to its truthfulness. A photo might be obviously fake; the character in a video might stand out as a creation. However, this stuff will get more complicated as we go along. 

We have mentioned this before, but one of our research themes deals with these types of issues. Part of the work will be technical; some of it will be more general relating to the American experiment and its possible contribution to age-old dilemmas that are becoming more troublesome. We have mentioned, several times, that we are building a portal. Many times, we add 'to truth' with this type of work in mind. 

For now, we will start pointers to material that is of importance plus some on-going commentary.  

  • TensorFlow - From 2019. We saw this contribution by Google to what is called deep learning earlier. And, we have read the discussions about this as being the new AI. The approach has gotten attention due to its demonstrated effectiveness to the extent of watching. These things used by this approach are mostly black boxes for which lots of effort is being placed with regard to understanding what is going on. However, right now, the particular interest for us is the approach's use of graphics. This is not GUI in focus. No, we are talking the very core of our modes for modeling reality and for basing decisions upon such models.  
  • GitHub - This is a link to the material related to the TensorFlow article. That is, one of the modern benefits is having project and code management available for team work via the cloud. Nowadays, you see lots of papers offering their data via this method encapsulated with the algorithms used for the data. We are using this facility for our portal work, somewhat. 
  • Medium - This is the media that provides access to the TensorFlow article. We place it here as an example of sites that provide support in various ways which are going to be important in the future. 
We have to touched upon some of this, in our work, so far: Content can be configuration. And, we will be doing more. We have the technology blog for specifics which will become more active. 

Remarks: Modified: 02/27/2021

02/27/2021 --  

No comments:

Post a Comment