CafeTran
How can we help you today?
CafeTran Espresso 2016 Ichiro - Update 14
translator.json.JSONException: A JSONObject text must begin with '{' at 1 [character 2 line 1]
Hi Amos,
It means that MyMemory online resource returns empty strings (nothing). Perhaps, they are updating something on their servers and it will start working again soon.
Igor
>> lowered fuzzy match calculation when the matched words are present in the segment in a different order.
Okay, let's us think about the implications of this.
1. Do not trust this tool when it highlights no differences, but the match rate is (far) below 100%.
2. In that case, do a visual check/comparison for yourself. Watch out!
Sorry to say that I've never seen this kind of CAT tool. I'm not complaining, but I'm almost disappointed.
One last word
I have recently been feeling the course of development is a little bit off the track (IMHO).
Which should come first, anyway: the enhancement of the core features (increased accuracy of TM/glossary matching [including this issue of word order difference and improvement of stem look-up), or the addition of "peripheral" features (at least to me; though I admit they are very useful to and very much welcome by other users)?
If you think CT's core features are already (near) perfect, pardon me. I'm living in a different world.
IMHO, a house built on a fragile foundation is destined to collapse sooner or later, though makeshift patchwork would keep it alive a little bit longer.
That's my one last word.
Cheers to you all.
P.S. Please don't waste your time to respond to this. 'Cause I'm not gonna follow this topic.
Michael Beijer
I think Slate Desktop needs to be tested by Hans Lenting, before we can make any kind of final judgements as to its quality. I haven't been able to produce much of value, but I think this is primarily due to the small size of my own personal TMs, and probably also the messy condition they are in.
If I had time, which I never really do, I could test it for those of you with large, high quality TMs. However, I don't have time. Also, who would want to hand over their life's work to a crazy Dutchman?
Michael Joseph Wdowiak Beijer, the 1st. Count of Hastings and its Environs.
Hi M,
The notion of "core features" is really relative. For some translators these are glossaries, others never use them and rely on translation memories. Another group of translators might treat web resources as core features, not to mention those who need the 'near perfect' support for projects coming from other CAT tools. CafeTran is updated quite frequently and in each update I meet some requests focusing on one or two areas of development. Improving the user interface has been a priority in CafeTran 2016 updates. The stem look-up based on Hunspell has also been introduced this year. The accuracy of prefix matching for fragments and segments in translation memories should be better than any other stem-based methods, especially for longer fragments. By the way, I wonder if any other CAT tool has it? The issue of word order difference in fuzzy matching has just been tackled in this update so the percentage accuracy is now lowered, which has been requested too. A colored highlighting of such word order difference will follow. The path of development is a mixture of the requests from numerous translators and the developer's ideas. I am currently working on Slate desktop integration, and yes it was requested too. However, I do understand the complaints if some requests are not met immediately. It usually means that I could not find the time or the feature is still under development. So the "little bit off the track" feeling may be a temporary one and in some future updates you may find what you are looking for.
Cheers,
Igor
IK: The notion of "core features" is really relative
It is. I, for one, think you should pay more attention to the KB (as promised), not only offering HOW TOs, but also WHYs.
I am currently working on Slate desktop integration
How many people actually use it? How much better is Slate as compared to CT's MT features? If it is better at all (which it should be, theoretically and in the light of the purchase price). I asked for a comparison quite a few times, never got it. The reviews I read haven't been very encouraging so far. What do you think?
H.
By the way, Tom committed a deadly sin. He misspelled CafeTran on his homepage:
And to add insult to injury, there's still no Mac version.
H.
>> lowered fuzzy match calculation when the matched words are present in the segment in a different order.
I confirmed that the rate is actually far lower than before (in one case, 79% vs 99%).
But if this is intended as a permanent solution, I wouldn't agree to it.
Cheers,
I just clicked on Resources -> Abbreviations -> Add selection to abbreviations. CafeTran closed (without saving). I haven't looked to see if it did add the abbreviation, but then I don't know how to as there's no documentation. (When I suggested setting up a wiki, Igor said I could set one up myself and have a link from the Knowledge Base page that no-one even knows exists.)
This sort of thing has to be priority - you can count the number of Slate users on the fingers of one hand, ditto TMTown, no-one cares about the tie-up with ProZ, but we all waste buckets of time rebooting, restarting and looking for workarounds for the innumerable bugs!! I no longer use the web resources due to the constant crashes and freezes. (I use autokey instead.) I would never recommend CT to anyone who didn't have at least a PhD in computer science and preferably was also a practicing Buddhist (due to the level of patience required).
Yours in constant frustration,
Jeremy
Hi Jeremy,
Probably, you use some outdated version of CafeTran where one of the web resources (Linguee.com) crashed CT. Some complex web resources with the heavy use of javascript can crash the Java web browser, which is rather a Java issue. The confirmed problem with the Linguee.com web resource has been fixed. I haven't received reports about other problematic web resources so far.
As for the " Add selection to abbreviations" issue, this is the first time the issue with this feature has been reported. Again, please update CafeTran to the latest release and submit the support ticket if this or other issues occur in the latest release as well. If confirmed, they get the high priority for the fix.
Igor
>But if this is intended as a permanent solution, I wouldn't agree to it.
How about some colour marking?
How about some marking colour?
>- translation consistency check - an issue in the similar segments with different numbers only and the Autopropagation option on. Such segments were falsely reported as inconsistent.
Yay!!! Thanks!!!
Igor Kmitowski
Hello,
CafeTran Espresso 2016 Ichiro - Update 14 is available for download. The update can be performed via the Drag & Drop procedure as follows:
1. Run CafeTran.
2. Download 20160809_update.zip file from here (do not unzip or rename the file after downloading) and place it on your desktop.
Note: On Mac OSX, the Safari web browser may unzip the file automatically after downloading. To prevent this, just use the right-click menu and choose the "Download Linked File" command.
3. Drag and drop the downloaded file into the Project Dashboard and restart the program.
4. Drag and drop the downloaded file into the Project Dashboard and restart the program. Second time!
The step 4 above is only for the users who missed the preview (Harbinger) and Ichiro updates with the new icon sets.
On Mac OS X, the Java 1.8.0_77 update broke the drag and drop function in CafeTran. The fix is available starting from CafeTran's Update 4. If you skipped all the updates from 4 to 13, please choose the downloaded update file via the Help > About > Install Update button instead of dragging and dropping it into the Project Dashboard.
The update 14 fixes and improves the following:
- lowered fuzzy match calculation when the matched words are present in the segment in a different order.
- translation consistency check - an issue in the similar segments with different numbers only and the Autopropagation option on. Such segments were falsely reported as inconsistent.
- splitting QA trailing and leading spaces check into two independent checks.
- embedded display of Google Maps based on the user's available location in the new ProZ.com "What I am working on" feature for translators.
- added an option in the Project Segments pane to set the Filter on the found word/phrase directly after performing the search via the Search bar.
- increased font size for the resource name field in the Autoassembly pane.
- another color for the assembled result in the Match bar to distinguish it easier from the TM fuzzy match results.
- a modified "Image background - bright" shaded theme.
Have a nice week!
Igor
2 people like this