What outdated & incorrect info have you found in the Wiki?

Have you found something that seems to be absolutely wrong in the wiki? If you’re not sure enough that it is actually wrong, post a link to it in this thread. Someone in the community should be able to say for certain.

Maybe it is obsolete information. Maybe it is a feature that has to be accessed in a particular way. Maybe the documentation is ambiguous.


Here is my Gramps bafflement for today.

Hopefully it will be the only confusion for the day.

The wiki says that a backup archive can be extracted to a fresh tree via the Manage Family Tree dialog.

I’ve not seen this and don’t know how a user could navigate to the folder with the archive. I suppose you could change the path of the database folder in the Preferences… but then you’d lose access to your other trees.

Is this an old feature that evolved into a different method? The How to restore a backup says its instructions are for v3.3 and later.

The Manage Family Trees sections says that this is only an option with the GNU Revision Control System. (The WHAT?)

If you are using windows then you won’t see the archive button which basically enable a snap shot (archive) feature.

Sorry link dump from googling!

Archiving a Family Tree and Gramps and Windows: Archiving which leads to a closed bug that shows another bug related to windows ( 3125 Any RCS activity causes an exception ) which lead to a discussion on reddit ( Looking for workflow/toolset for storing GrampsXML in online git ) which leads to a feature request to replace RCS with something more modern that windows can use ( 4901 : Add a new storage format base on GIT) !

Also Database_Formats#Versions and old message archiving-problem which may answer your question as someone called Doug mentions about RCS archives:

It organises the backups in a logical tree structure so it’s immediately
obvious which backup comes from which version of the DB and when, and
which you’re loading if you need to use an earlier backup. It saves
having to think how you might name your backups to show their
inter-relationships and then remembering what was the system you used
the year before last or whenever, when the time comes and you need it.

1 Like

That’s the 4th thing I’ve heard the Windoze version of Gramps can’t manage yet is doable in the other OSes.

The others were:

  • remote host and port support for the locating the database remotely
  • network paths without mapping a drive letter assignment
  • environment variables

any more?

this is wrong, it’s no problem connecting to a database on a remote computer as long as you actually use a database engine that support host name and ports.
sqlite do not, it is an in prosess database engine.

I have multiple databases both running on postgresql and mongodb servers that connect with host name or IP address and port number

it is also possible to do so from wsl, wsl2 and from any virtual client.
It is also possible to connect to a server on wsl, wsl2 or any virtual client…

only problem was that there was a change in how it was done from wsl to wsl2 how to connect to the host os.

Thank you. Such clarifications are welcome.

It also helps people understand the reasoning behind previous inquiries about the future of the MongoDB and PostgreSQL add-ons.

I have not tried this for Gramps 5.x, but if it doesn’t work, it is only because Gramps doesn’t support UNC file path, Linux use i.e Samba for this.

But there is many Windows desktop software in general that don’t support UNC, not unique for Gramps, an actually, storing a sqlite database on a network share for read/write is not recommended because of latency, even though sqlite actually support it, regardless if it is a mapped network share or an UNC share. a sqlite database should be stored on the same system the software is run from, not needed to be same drive, just the same system…

I do not see any huge differences storing the database file on my host computer and accessing it from a virtual client on the same computer, but I do see reduction in performance on larger database files stored on network shares vs. local computer, even when the share is fast nvme ssd over 2GB (1+1) paired network.

The latency differences for the same database stored in posgresql remote or local are minimal.
The same is it if the postgresql database is installed on Linux or Wndows, and similar hardware, no noticeable difference.

mongodb drivers needs updates before its any point to do any more speed tests on it.

It should have been interesting to see how Gramps would have performed as graph database software…

The ‘Start with Genealogy’ page refers to procedures and UI elements in Gramps that don’t match with what the user sees in v5.1.

I’m new to genealogy and Gramps, and tried using the wiki as a gently introduction to Gramps and the general concepts involved, but the difference between them makes it too confusing or unclear for me to confidently follow as a novice.

If it will eventually be updated, in the meantime it might be ideal to warn readers that some of the information is not current and might not match their version of Gramps.

Sorry that you’re having problems.

Procedures that vary from 5.1? Can you give us an example?

I see you’re in macOS. The majority of the screen captures are from a Windows computer using the default “Adwaita” theme in English. Some are from various flavors of Linux boxes using different Desktops.

Because of theming, GUI elements have a wide variety of appearances. You might have customized your OS with a different theme and Gramps would be affected.

Could you post a screen capture of the main screen on your computer?

I see a lot of development pages that are quite old, and mention tools that were modern at the time, but not anymore, like Eclipse. And this one is particularly old, because it still refers to 4.2 and good old 3.4:

https://gramps-project.org/wiki/index.php/Running_a_development_version_of_Gramps

That 3.4 is nice if you really like to work with a flat place database, like I did for quite a long time, but the 4.2 section should be updated to 5.1, including references to current versions of Debian, Fedora, and ubuntu.

Hi, thanks for the reply.

Revisiting the issue (and the wiki), I think it might be a misunderstanding on my part. When I’ve read some of the steps, I’ve been trying to find the elements described in the window I’m working in, but it seems the guide is referring to other windows without explicitly telling me to switch between them. For example, in the Time for your first Person section, I’ve opened the Person editor to add a new person, per paragraph 3. Going on to the subsequent paragraph, I’m still in the Person editor, but I’m not seeing the things it’s referring to, e.g.

As you enter data, you need to add where you obtained this data (Source). For this, click on the Sources tab, then open the Clipboard from the Toolbar,

Because it I was looking for those things in the Person editor and the paragraph didn’t mention to switch to the main window, it seemed that there was a disparity. Only after playing with Gramps a little more and re-reading the guide did it make a bit more sense.

In my case, the steps could be a bit more specific. Screenshots might have helped, too.

Otherwise, there seems to be some minor inconsistencies in terms used. For example, in the same section:

In the Navigator, click on the Person icon to see the person view

and

For this, click on the Sources tab

is a bit confusing. The Navigator only shows a People icon, but is referred to as a ‘Person’ icon. There’s reference to a Sources ‘tab’, which seems to be the same kind of element as the People/Person icon previously mentioned.

1 Like