Let me see if I can be a bit more constructive than advising to avoid Across altogether, here. The worst CAT-tool is certainly not Across,I have seen worse, although the satanic desire to burn the tool and its users is becoming cumbersome - to say the least, let me remind some users that when posting in this forum, this is the sentence you see in the box above your comment: "This forum is dedicated to providing and receiving support on software." Let's stick to that, please, and vent your frustrations elsewhere.
Claudia Mirza wrote:
We are looking to move our production team to only Across
Although from a production point of view an understandable avenue to explore, this seems always a bad idea in my eyes, regardless of the CAT-tool you choose: why limit your resources to only a group of translators based on tool knowledge instead of actual translation knowledge?
1. We work a lot with PHI material that must be legally kept in the US. We know there is a way to mask certain text in documents going through across, but if you have documents that basically all have the same layout with information such as names, addresses etc. that should be removed is there a way to set up a template that knows to strip out certain data? We want to make sure if we have to sent PHI documents out of the US for translation, there is no personal information that can be viewed by the team working on it.
If a document must be legally kept in the US, it seems to me that uploading it onto a server from where non-US based translators can access it, would actually constitute a breach of that legal demand and therefore should be avoided - but I am no lawyer. The best way forward would in my opinion be to make that template yourself, have that translated and insert the various names, addresses and other proprietary information on your end.
2. CrossWeb was always painfully slow outside of the US for people working with it. Has anyone had a good solution to this problem?
Nope, but I use Across as a stand-alone solution.
3. When importing existing TMX files into Across, what is the best process to make sure all the text is moved and is specifically associated with one client? Before it seemed like when we would import TMX files and assign a Relation and Subject, it would not segregate the text properly.
That may have to do with the origin of the TMX files. Contrary to popular belief - that TMX is a universal standard - I have the distinct impression (I may be wrong) that the various CAT-tool producers cannot resist adding their own minuscule variations to that standard because it is not in their interest to have a universal standard, they want to lock users into their own tool as much as possible, which makes sense from their point of view. If it really was as universal as claimed, then a TMX file generated with Trados could be uploaded glitch-free in Wordfast Anywhere or Across (I can assure you that in these two cases, they usually lead to glitches and extra work). Likewise, I have supplied agencies with Across-generated and Wordfast Anywhere-generated TMX files that they uploaded in their Trados environment, and they ran into similar issues as you describe. I have found that advising them that they could align the source and target text in their own CAT-tool environment as they see fit for the best alignment is usually the best way forward.
If anyone has any feedback it would be much appreciated!
From what little information I can deduct, it seems you are looking for a way to standardize the way you collaborate with your translators. Why not get yourself a CAT-tool that you are comfortable with and dedicate staff internally to handle the source texts and target texts the way you see fit? That way, you allow yourself much more freedom in finding the translators you need for a particular assignment.