>The Grid and the Translation Editor take both 50 % of the width of the GUI.
A fresh CafeTran will then have 4 quadrants, which is very easy for referral during training. Besides that: migrators often say that CafeTran's Grid is so small (a peep-hole).
No time to comment on the rest (have to work), but I think your suggested Jump settings are dangerous. They will mean all kinds of incorrect stuff might be jumped right over. I generally have pretty much none of them selected, and carefully check each segment as I go down.
I do think the default Alt+Down should set the finished segment to Checked (i.e. highlighted grey). That is, this one should be the default: "Add checked segment to memory and go to next unchecked segment"
>but I think your suggested Jump settings are dangerous.
Very well possible. It's a discussion paper.
I still do not get the point about "Characters for removal". I tend to empty this.
The window size and the RAM size could also be part of an introduction during the install or (perhaps better) during the first start. Somehting like:
This is is the first time you are using CafeTran. Before starting the app, we would like you to make two [or more] settings. You can change these settings at any time under [Menu path to Settings/Preferences, which depends on OS].
and so on (beware of my German-Pidgeon English). Sounds childish, but many newbies like this. IMHO any preset RAM > 1 RAM might cause problems.
There are probably other settings to discuss. I just remembered these ones, for opening TMX files (e.g. for migrators). Now:
Wouldn't it be better to change this default behaviour to:
Your input please? I know settings like Fuzzy and Hits and Match case are language-dependent, but shouldn't CafeTran go for a subset of all languages? And, if so, for which one?
In the Dashboard, Project glossary should be ticked.
These ones should be ticked too:
And this is the 4 quadrant layout I was referring to (here on a 13" MacBook Pro, in full screen mode):
BTW: Regarding the RAM, I noticed that Igor's already set that to 2048 by default. A wise decision, me thinks.
I have this strange feeling that Igor knows what he’s doing, and that the current default settings reflect the way he thinks how they should be for most translators. His “philosophy” so to say. A philosophy we unfortunately shall never become acquainted with, because he refuses to write a manual, and there isn’t much progress in casu KB entries (promised to be ready by the end of this year) either.
That said, I can imagine it’d be useful if the download of the free/trial version of CT would be accompanied by a ReadMe file that explains the settings for source and target file essential settings like Segmentation (SL) and where to save any Hunspell dics (TL). Or even make available a first "download on request," with the settings set according to the specifications of the newbie.
Most of the settings suggestions mentioned above are highly individual choices, and should remain the way they are.
I don't think all users think all the features you think are terrific are terrific. I for one disagree with most of them, and the one(s) I do agree with shouldn't be changed either, as far as I'm considered. It's Igor's tool, and Igor Knows Best. I don't agree with that either, let's change that into Igor Codes Best.
Everybody thinks his suggestions are the best, so I'd say, write a ReadMe with your settings suggestions with links to the Wiki and ask Igor to include it in the download ZIP, or at least refer to it on the download page, and include in that ReadMe suggestions for the setting for "deviant" language pairs. The latter really will have to be changed for CT to work properly.
To some extent Woorden is right. This is what my proposal was for - to create a kind of "take a tour (and make perhaps some settings)" introduction that starts after the first installation of CT (and that can be skipped of course).
Readme-Files are hardly ever read, even by advanced users.
Another thing (and I do not exclude myself from contributing) is to have a dedicated 'Best Practices' or 'Learn from the Pros' section in the wiki - this has not been done yet. Such a section could contain several subsections for different languages and approaches. Approach means here e.g. "working mostly with AA" (Woorden), "working with big glossaries" and so on, with a clear explanation why certain settings should be activated/deactivated and what consequences these can have. Maybe we could concentrate on this and let Igor make his own decisions (as he always does, of course).
I must confess that I regularly stumble upon different settings and functions I did not know or use before. It is - due to the sophisticated complexity of CT - hard to become a kind of advanced user (a "CafeTran Pro", hehe) and to convey your knowledge and your experience
From this thread I conclude that Fuzzy & Hits isn't the optimal DEFAULT setting for TMs. But of course, this is open to discussion.
Hans CafeTran Wiki
I'd like to start a discussion about optimal settings of a plain Vanilla CafeTran (a fresh CafeTran installation) for a novice user.
I think that some of these settings will be language-dependent but nevertheless we should discuss these settings.
I'd also like to suggest a slightly different layout for a fresh CafeTran:
A predefined set for trimming new terms?
More RAM by default? What's the average amount of RAM for new computers? 4 GB? 8 GB?