| Pages in topic: [1 2] > | Creating a Simpler CAT Application for Everyone Thread poster: Zelikazi
|
|---|
Zelikazi
United States
Local time: 00:54
Japanese to English
+ ...
I think there are many translators out there who would like to utilize CAT software, but find the existing options to be difficult to use, or to conflict with their workflow.
I'm a JA-EN translator/web developer working on a simple alternative, and I'm looking for suggestions.
I don't intend to replace the established applications overflowing with features that are already out there.
Rather, I want to build a simple, modern CAT application that gives us CAT featur... See more I think there are many translators out there who would like to utilize CAT software, but find the existing options to be difficult to use, or to conflict with their workflow.
I'm a JA-EN translator/web developer working on a simple alternative, and I'm looking for suggestions.
I don't intend to replace the established applications overflowing with features that are already out there.
Rather, I want to build a simple, modern CAT application that gives us CAT features without forcing us to do things a certain way.
Some of the features it will have:
- Looks up terms in multiple dictionaries/sites in advance for the translator
- Autocompletes lengthy terms (optional)
- Supports translation memories, term bases, etc.
- Autosaves work
- Online and accessible from anywhere
- Support for many language pairs and common document types
- File encryption and secure connections
- No clutter, no bloated software
I'd like to hear what sort of application and what sort of features fellow translators would like to see!!
Any ideas?? ▲ Collapse
| | | | Roy OConnor (X)
Local time: 06:54
German to English
You list of features sounds pretty comprehensive to me and I'm left wondering what you have left out to stop it being another piece of "bloated" software.
| | | | Samuel Murray 
Netherlands
Local time: 06:54
Member (2006)
English to Afrikaans
+ ...
Zelikazi wrote:
I'm a JA-EN translator/web developer working on a simple alternative, and I'm looking for suggestions. ... I want to build a simple, modern CAT application that gives us CAT features without forcing us to do things a certain way.
I'm afraid all CAT tools force you to do things a certain way. For example, some CAT tools use online components that force us to use the web-browser for things that we would otherwise normally have used a word-processor for. (-: And to make matters worse, those features often work only in one or two specific browsers or versions of browsers.
Some of the features it will have:
- Looks up terms in multiple dictionaries/sites in advance for the translator
A useless feature, I think. It is often not possible to predict which "terms" the translator will have problems with and would want assistance with, and if you simply try to look up all possible terms or combinations of words, you're going to end up with lots of processing power used up for no benefit. Being able to look up terms is useful, I think, but the translator has to initiate the look-up himself (or the function to determine which terms the translator might want to look up has to be super intelligent, so as to look up only so many terms that are necessary).
- Autocompletes lengthy terms (optional)
This should not be optional. Various types of auto- and semi-auto-completion is what makes a CAT tool useful for the translator.
- Supports translation memories, term bases, etc.
I'm not sure what you mean. Would it be a "CAT tool" if it doesn't have this feature?
- Online and accessible from anywhere
Offline is better, faster, easier, etc. Online is slower. What do you mean by "anywhere"? The ability to access your work via a smartphone or tablet may be a nice gimmick, but no translator will do actual translation via such devices. Optionally do what Idiom did -- use the online system as the project management side, and provide the translator with both an online and an offline translation environment to choose from.
- Support for many language pairs and common document types
I'm not sure what you mean by "many language pairs"... isn't that the easiest part of the development? But "common document types" is the thing that may well kill your project, because the most common document formats are not open and will need to be reverse engineered.
I'd like to hear what sort of application and what sort of features fellow translators would like to see!
Different translators have different needs and preferences.
Bored? Try this: fork Pootle (it's GPL) and give it the following features:
1. Ability for registered user to upload his own file for translation (and download both the XLF and the final format of the translation).
2. Ability for user to upload and/or connect to his own translation memories and glossaries
3. Ability to upload (and translate) files in common formats (e.g. DOCX) instead of having to convert them to XLF manually with some third-party tool.
4. Global find/replace.
5. Ability to keep a version of the translation table in a separate window that contains *all* segments, while translating in the usual window that contains only ten segments.
6. Segment statuses (and ability to set statuses of multiple segments in one action).
7. Ability to disable use of free-and-open translation memories and using only one's own.
8. Ability for a new user to register a project-wide (not site-wide) "administrator" account, so that he can be in full control of his own translation space on the server, including e.g. ability to add users and give them permissions.
[Edited at 2014-12-12 11:45 GMT]
| | | | | Sounds like another version of the old story | Dec 12, 2014 |
OK, different tools and different languages have different needs.
Trados was getting too big for some and its advertising was a pain, so Yves Champollion came up with Wordfast, now Wordfast Classic.
http://en.wikipedia.org/wiki/Wordfast
I used it myself for a time, but came back to Trados.
Then everything went orange, and Istvan Lengyel & co, aka Kilgray,... See more OK, different tools and different languages have different needs.
Trados was getting too big for some and its advertising was a pain, so Yves Champollion came up with Wordfast, now Wordfast Classic.
http://en.wikipedia.org/wiki/Wordfast
I used it myself for a time, but came back to Trados.
Then everything went orange, and Istvan Lengyel & co, aka Kilgray, presented us with MemoQ.
They wanted something 'translator-friendly' and are providing some very keen competition to SDL.
But to do so and fulfil the translator's real needs, they have also come up in the powerful league, and Champollion's team had to create Wordfast Pro to meet demands for more functionality.
I got tired of Wordfast Classic because it was too simple.
Basically, if you want the support and the features you have to spend time learning to use them. For better or worse, the software cannot read your mind, but it can be made to do largely what you choose. It just gets complex... and developers need to be paid for their efforts too, so it costs money! ▲ Collapse
| | |
|
|
|
Heinrich Pesch
Finland
Local time: 07:54
Member (2003)
Finnish to German
+ ...
| Which tools have you tried? | Dec 12, 2014 |
There are a lot of different tools on the market, some new and rather simple, some old and with lots of functions. Your list of functions means a rather complicated piece of software, which would take years to compile and test. Why invent the wheel again? Software development in our era is not a one person's job anymore. You'd need a team.
| | | | | Won't work as a business venture | Dec 12, 2014 |
Other people have said it, but it bears repeating: developing a CAT tool is a large and complex job, hardly for one person and hardly as a side project. I suspect that implementing the import and export of office formats alone would be a major job in itself. Providing support could easily prove to be a full-time job as well, if you end up having a decent number of users. And if you want to charge for access, you'll be competing with a free product from an established competitor: wordfast anywher... See more Other people have said it, but it bears repeating: developing a CAT tool is a large and complex job, hardly for one person and hardly as a side project. I suspect that implementing the import and export of office formats alone would be a major job in itself. Providing support could easily prove to be a full-time job as well, if you end up having a decent number of users. And if you want to charge for access, you'll be competing with a free product from an established competitor: wordfast anywhere. You would have to produce something extraordinary in order to be successful.
Also, I hate in-browser CATs. I don't think it's a good idea, and most professional translators seem to agree, so the odds are stacked against you on that front as well.
I agree with Samuel: if you want to mess with CAT tools as a hobby activity, contribute to some open source project. I'd suggest an offline tool, perhaps OmegaT. If you don't want to be integrated into/subordinated to a project, you can fork it, or develop somewhat independent features/modules.
If you're looking for a moneymaking opportunity, perhaps try something other than a full-on CAT tool.
I'm somewhat in the same boat as you: I wrote an aligner and a TM lookup tool. Both are free & open source and both are hobby projects, but both contributed to some sort of income in some fashion, directly or indirectly. ▲ Collapse
| | | | Samuel Murray
Netherlands
Local time: 06:54
Member (2006)
English to Afrikaans
+ ...
| Online version of OmegaT | Dec 12, 2014 |
FarkasAndras wrote:
If you want to mess with CAT tools as a hobby activity, contribute to some open source project. I'd suggest an offline tool, perhaps OmegaT.
If you want something web-based, start by checking if you can get the online version of OmegaT running. It's called Boltran and the files are on sourceforge in Git.
If you're looking for a moneymaking opportunity, perhaps try something other than a full-on CAT tool.
It did not even occur to me that you might want to make money (-: well, do you? If you want to do that, then perhaps an easy way of doing that might be to host an existing tool and charge people money for it.
If I understand the GPL correctly, you can legally host your own custom installation (with customisations) of both Boltrans and Pootle, for money, even if you have integrated your own enhancements with it, without having to share it with "the community".
| | | | Zelikazi
United States
Local time: 00:54
Japanese to English
+ ...
TOPIC STARTER
Thanks to all of you for providing so much feedback. Your responses are helping me to frame the context in which other translators view this idea. Though most responses have not been especially encouraging, negative feedback is just as important as positive feedback and I hope to integrate these opinions to manage my own expectations.
Individual responses will follow.
| | |
|
|
|
Zelikazi
United States
Local time: 00:54
Japanese to English
+ ...
TOPIC STARTER
Roy OConnor wrote:
You list of features sounds pretty comprehensive to me and I'm left wondering what you have left out to stop it being another piece of "bloated" software.
Thanks for giving me an idea of where I should stop. I'll keep that in mind.
| | | | Zelikazi
United States
Local time: 00:54
Japanese to English
+ ...
TOPIC STARTER
Samuel Murray wrote:
I'm afraid all CAT tools force you to do things a certain way. For example, some CAT tools use online components that force us to use the web-browser for things that we would otherwise normally have used a word-processor for. (-: And to make matters worse, those features often work only in one or two specific browsers or versions of browsers.
This is true. My hope is to make the application as non-intrusive as possible by functioning very similarly to a word processor and eschewing the forced line-by-line translation method of existing CAT applications (some people will like this, some people won't). Also, cross-browser compatibility will be a given.
Some of the features it will have:
- Looks up terms in multiple dictionaries/sites in advance for the translator A useless feature, I think. It is often not possible to predict which "terms" the translator will have problems with and would want assistance with, and if you simply try to look up all possible terms or combinations of words, you're going to end up with lots of processing power used up for no benefit. Being able to look up terms is useful, I think, but the translator has to initiate the look-up himself (or the function to determine which terms the translator might want to look up has to be super intelligent, so as to look up only so many terms that are necessary).
I am working on making it intelligent with matching terms (matching phrases, ignoring extremely common words, etc.), but also allowing the translator to initiate his/her own simultaneous look-ups quickly if they are looking for something else. This is a very difficult problem, though.
Offline is better, faster, easier, etc. Online is slower. What do you mean by "anywhere"? The ability to access your work via a smartphone or tablet may be a nice gimmick, but no translator will do actual translation via such devices. Optionally do what Idiom did -- use the online system as the project management side, and provide the translator with both an online and an offline translation environment to choose from.
Only word look-ups from other sites would require being online, the rest would all be done client side and be instantaneous. By "anywhere", I intended to mean multiple computers that may or may not have translation software installed on them. As for a completely offline option... that would be very good, but will require further exploration on my part.
But "common document types" is the thing that may well kill your project, because the most common document formats are not open and will need to be reverse engineered.
This is something I am concerned about. My clients tend to stick with Microsoft Office formats almost exclusively, so at least the formats are well-known and there's not too many of them, but other translators might encounter different formats.
Bored? Try this: fork Pootle (it's GPL) and give it the following features:
I'm going to take a good look at this (although I don't use Python). It seems intriguing.
| | | | Zelikazi
United States
Local time: 00:54
Japanese to English
+ ...
TOPIC STARTER
Sounds like another version of the old story
The translation tools do all seem to run together after a while...
The biggest difference with this would be that it won't enforce line-by-line translating, dispensing of the related formatting and workflow structure. Obviously many translators won't want that, but they already have many options available (as you've said).
| | | | Zelikazi
United States
Local time: 00:54
Japanese to English
+ ...
TOPIC STARTER
Which tools have you tried?
I've been searching for CAT tools that provide lookups, autocomplete, and (maybe) term lists without requiring a certain structure or format. I haven't found any yet, but I'm all ears.
Software development is difficult, but web development tools have advanced considerably in the past few years. I think a beta version for certain language pairs will be feasible without a big team.
| | |
|
|
|
Zelikazi
United States
Local time: 00:54
Japanese to English
+ ...
TOPIC STARTER
FarkasAndras wrote: Other people have said it, but it bears repeating: developing a CAT tool is a large and complex job, hardly for one person and hardly as a side project. I suspect that implementing the import and export of office formats alone would be a major job in itself. Providing support could easily prove to be a full-time job as well, if you end up having a decent number of users. And if you want to charge for access, you'll be competing with a free product from an established competitor: wordfast anywhere. You would have to produce something extraordinary in order to be successful. Also, I hate in-browser CATs. I don't think it's a good idea, and most professional translators seem to agree, so the odds are stacked against you on that front as well. I agree with Samuel: if you want to mess with CAT tools as a hobby activity, contribute to some open source project. I'd suggest an offline tool, perhaps OmegaT. If you don't want to be integrated into/subordinated to a project, you can fork it, or develop somewhat independent features/modules. If you're looking for a moneymaking opportunity, perhaps try something other than a full-on CAT tool. I'm somewhat in the same boat as you: I wrote an aligner and a TM lookup tool. Both are free & open source and both are hobby projects, but both contributed to some sort of income in some fashion, directly or indirectly.
It is a complex task. I think you are right about the difficulty with office formats. My hope was that by building an application that I would use (as a translator who usually receives work that doesn't lend itself to the line-by-line format and who would appreciate more freedom and simplicity), that maybe I could add something to what's out there using modern web development techniques.
Thanks for showing the tools you developed to me, as well.
I would like to hear about why you and others dislike in-browser CATs. Could you tell me more about that?
| | | | | in-browser CATs | Dec 13, 2014 |
It just doesn't work well IMO. First of all, they are bound to be slow if your internet connection is wonky, and if your internet goes down or you're in an area where there is no internet (e.g. on a plane) you're SOL. Even if you have a good connection yourself the site itself may be slow and the servers may have a bad hair day any time, usually when a deadline is coming up.
Then there's the issue of TMs and termbases. Do I seriously want to upload all my data for every project? No. I have... See more It just doesn't work well IMO. First of all, they are bound to be slow if your internet connection is wonky, and if your internet goes down or you're in an area where there is no internet (e.g. on a plane) you're SOL. Even if you have a good connection yourself the site itself may be slow and the servers may have a bad hair day any time, usually when a deadline is coming up.
Then there's the issue of TMs and termbases. Do I seriously want to upload all my data for every project? No. I have a lot of this stuff and it'd be a pain to upload it all and manage it on someone else's servers.
In-browser CATs are usually the idea of companies (clients) that want to control the whole translation process themselves. "Hey, let's not let these pesky freelance translators make their own project and QA settings and segment their own files and select their own reference material. We'll take care of all of that, we know better! And we certainly won't let them just download all our TMs. They might use them for other projects then!" Of course this is an area where the perceived interests of the two parties are opposed. Freelancers usually want to manage their work themselves, that's why they are freelancers in the first place. ▲ Collapse
| | | | | Browser-based tools | Dec 14, 2014 |
Zelikazi wrote:
I would like to hear about why you and others dislike in-browser CATs.
Don't assume everyone is against browser-based CAT tools just because you saw posts by a couple of people who dislike them. Thousands of people are using Wordfast Anywhere, which is entirely browser-based, and most seem to be happy with it.
Now the question is: what would your own browser-based tool bring to the party that Wordfast Anywhere isn't offering already, especially if you intend to ask money for your would-be tool (given that Wordfast Anywhere is free)?
| | | | | Pages in topic: [1 2] > | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Creating a Simpler CAT Application for Everyone | Protemos translation business management system | Create your account in minutes, and start working! 3-month trial for agencies, and free for freelancers!
The system lets you keep client/vendor database, with contacts and rates, manage projects and assign jobs to vendors, issue invoices, track payments, store and manage project files, generate business reports on turnover profit per client/manager etc.
More info » |
| | Trados Studio 2022 Freelance | The leading translation software used by over 270,000 translators.
Designed with your feedback in mind, Trados Studio 2022 delivers an unrivalled, powerful desktop
and cloud solution, empowering you to work in the most efficient and cost-effective way.
More info » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
Translation news related to CAT tools
