My dream CAT tool…

… probably hasn’t been created so far, but as I have recently learned to use several tools such as Google Translator Toolkit and Translation Workspace I started wondering what features it should have to make translation process easier, more effective, less time-consuming and stressful. So I decided to create my list of must-have features which might inspire CAT software developers if they ever happen to visit my website 🙂

1. Vertical (column) view with the original text on the left and translation on the right. I personally hate horizontal views which were used for instance in Trados Translator’s Workbench and Wordfast and have been adopted in Translation Workspace. Well, the first CAT tool I have ever used was memoQ with a column view, so I simply got used to that way of presentation, but in my opinion it is also much easier to compare source and target text when the equivalent phrases are more or less on the same level (although it is not always the case, as it is sometimes necessary to restructure the sentence taking into account the theme-rheme requirements of a given pair of languages).
2. Easy application of formatting. I prefer when the text is totally stripped of formatting and I don’t see strange colours, fonts etc. in my CAT tool. Of course, I should be able to use bold, italics and insert formatting tags where necessary, but I don’t want to see anything I don’t need to.
3. Adjustable tag view. Sometimes it is necessary to put tags in the right order so that the target text is formatted appropriately. To do that I need to see at least a basic description of the type of tag I am inserting (is it bold, change of font or whatever?). However, I should be able to minimize lengthy tags descriptions for example when I want to scroll through the text quickly or just re-read it without checking tags. And this option should be easily available and not hidden somewhere in a huge general settings menu.
4. Standard or at least customizable keyboard shortcuts. I am always impressed by the creativity of software developers when it comes to creating new keyboard shortcuts for confirming a segment or moving to a next one. Why CTRL + Enter can’t become a standard just as it is with CTRL + C for copying or CTRL + V for pasting things? It is comfortable, I don’t need to use both hands (which means I can hold my tee cup with the other hand), so why create some strange, finger-cramping combinations such as CTRL + ALT + +, etc.? Well, it can be argued that in these types of CAT tools in which you don’t see the whole text at one time (usually with horizontal view) you need to have keyboard shortcuts for moving forwards and backwards. I would also standardize shortcuts for spell-checking (F7), quality check (F8), saving (if it is not done automatically), concordance (CTRL + C) and so on. It would be much easier to jump from one CAT tool to another and keep our effectiveness level high.
5. Autosave. One may say that if we attach a translation memory to the project (and we usually do), it is not necessary for the CAT tool to save our progress. However, the thing is that if my computer crashes or I accidentally close the programme, I need to pretranslate the file or once again go through already translated segments, confirm 100% hits from the memory and continue my work. Apart from the fact that it is a waste of time (and time is money in translation business :)) I loose information about original match rate with a segment stored in the translation memory (e.g. provided by the client). And this is crucial if I weren’t supposed to change anything in segments with let’s say higher than 99% match rate. By the end of translation, I won’t remember which segments were which.
6. Predictive typing. Another useful feature connected with terminological databases. When you start typing a term which is in your database, the programme inserts it automatically so that you don’t have to type the whole word or phrase. In Studio 2011 there is an additional feature called Autosuggest dictionary which functions very similarly but the translator has to create a dictionary from a translation memory. Then you can insert not only typical terminological units but also pretty common phrases. However, its big disadvantage is that you have to have a translation memory consisting of at least 25 000 translation units to create an autosuggest dictionary.
7. Sorting and filtering functions. In my opinion a good CAT tool should have extensive sorting and filtering functions apart from a default “Find and replace” window. Sometimes it is much more effective to translate segments sorted in a particular way, e.g. starting from numbers, fuzzy matches etc. Apart from that when you have a file with XXk words and you can’t send it back to the project manager until it is 100% done and you see a message that there is still 1% to be translated/confirmed but you can’t see where this &*#(&@(* segment is, you don’t want to waste your time scrolling through the whole text.
8. Underlining spelling mistakes with spelling suggestions. It might seem that such a function should be a basic feature of CAT tools which are in a sense slightly more sophisticated word processors. However, it is not so common even though it makes translation and review much quicker.
9. QA checks. Unquestionably, it is a must-have! Some localization companies create their own software for QA checks, but to be honest it would be less time-consuming and easier to use in practice if CAT tools were equipped with good QA modules. Many of them already are, but they also generate many false warnings especially in the case of numbers.
10. Simple and user-friendly interface. I don’t suggest getting rid of many useful options and settings, but when I need to do a simple thing like translating a small file, I want to have necessary functions at hand, go through a quick wizard and start the actual translation instead of configuring hundreds of settings, looking through directories to find TM, TD and the source file, typing in various server addresses, attaching resources and so on. Nevertheless, if I become an advanced user or want to customize the tool to suit my very special needs, I would like to be able to do that by simply accessing advance settings menu or something like this.
11. Easy to configure segmentation rules. Several blog posts ago I described how to configure memoQ to divide segments after a semicolon. Well, it was fairly easy but so far I haven’t found a solution how to configure other types of segmentation (for instance after a number followed by a dot and a tab). You need to have at least some basic knowledge about regular expressions to be able to configure such advanced settings. I think it would be good to be able to “teach” the CAT programme our segmentation rules on the basis of past projects or at least download additional settings from the website of the producer.
12. Termbase and memory alignment tools embedded in the CAT tool. Well, I don’t align documents every day, but I don’t like switching between programmes when I want to create a new termbase or import to it some new terminology units.
13. Embedded Pomodoro function or other motivational features 🙂 I use Pomodoro almost every day so a CAT tool with time counter telling me when to have a break or reminding me of work to be done (warning activating after a set period of inactivity or a browser add-in blocking certain time-consuming sites) and presenting me some statistics after each work session would be very functional. I don’t know whether you do the same but I like counting how many pages/words I have translated per hour, etc. It’s a kind of a personal challenge and a way to forecast how much time I will need to finish the current project. Certainly, adding such simple feature would be a great advantage to me.

That would be all from me. And what would you expect from a dream CAT tool? Should it be able to react to voice, brew coffee, tell jokes, have embedded MP3 player or simply not crash every other day, work fast and so on? Feel free to add your ideas and comment mine here or using other social channels. You can also leave your vote below! Maybe someone will hear us 🙂