Words as Symbols in Software

[Mark Meer]

We have some (very simple) software that uses some simple English words, and we'd like to avoid having to translate the software if possible.

I'm trying to make the argument that words such as "ON", "OFF" or "MODE" that appear in the software can be just considered as symbols, who's functions are described in the accompanying user instruction documents (which are translated).

Is this a reasonable approach to not having to translate these terms in the software?

 

[Pancho]

encendido, apagado, modo.

Do the above look like symbols? Imho, words in a language I don't understand don't look like symbols at all.

For on-off, a better option is 1 and 0. For "mode", perhaps you could use symbols for each individual mode, if available. There are sets of universal symbols for safety, traffic, laundry care, etc.

Perhaps it's possible to find some real symbols, like icons or colors, that are more universal than English language words. That way you wouldn't have to translate, but still convey the meaning you want in almost any language.

 

[Mark Meer]

We use symbols wherever possible, but there are several concerns:

1. Although many symbols may be "universal" (e.g. from an ISO standard), that still doesn't mean that users know what they mean. They have to be explained in the instructions (IFU) regardless. So, for example, for a French market, what's the practical difference between the IFU stating:
a) "do not wash" ISO 3728 symbol = Ne pas laver
b) DO NOT WASH = Ne pas laver
c) Some symbol we made up = Ne pas laver

Obviously, in practice, wherever applicable we go with (a). But in other cases where there is no universal symbol, I don't see any practical difference between (b) & (c)...

2. There are many symbols that are so ubiquitous, but don't have an ISO or other standard to back up their "universality" (as far as I know...). Examples are a check mark, an X, the audio on/off symbol (see attached). These still warrant explanation in the IFU, so again practically speaking I don't see any difference between the IFU saying "X = non" or "NO = non"...

 

[MIREGMGR]

Is this a reasonable approach to not having to translate these terms in the software?

I'm not sure what you mean by "reasonable". I'll bet, though, that it won't fly regulatorily. An independent reviewer would say that the words are text. They're in one EU language. The EU national language laws are clear. Pretty much, end of discussion.

 

[Mark Meer]

Pretty much, end of discussion.

From unreasonable regulators, probably you're right. ...but by this thread, I'm also interested in a philosophical discussion, which I don't believe is so cut-and-dry.

Take the EN 980 "LOT" symbol for example. Apparently somewhere along the line regulators accepted this as a symbol, despite the fact that it is clearly an English word (albeit surrounded by a rectangle). If there were serious objections to what I'm proposing, they presumably would have chosen a different symbol.

 

[MIREGMGR]

Take the EN 980 "LOT" symbol for example. Apparently somewhere along the line regulators accepted this as a symbol, despite the fact that it is clearly an English word (albeit surrounded by a rectangle). If there were serious objections to what I'm proposing, they presumably would have chosen a different symbol.

The important distinction is that your LOT-in-a-box example (somehow) went through an international standard creation process and was accepted. Your proposed usages haven't.

Most regulators accept standards at face value. That doesn't mean that whatever you suggest should be considered by regulators to be logically equivalent to a standardized usage, will be accepted as such by those regulators. They might not agree with your logic, or they might feel constrained by what they understand to be a conflicting rule or law.

Whether or not you can convince your regulators to accept your justification in the short run, you certainly could pursue it for the long run. If you were to get the appropriate ISO technical committee to accept ON-in-a-box, OFF-in-a-box and MODE-in-a-box for inclusion in the next edition of 15223, then you'd be good to go.

By the way, "LOT" came into English from Dutch, and originated in Old German.

 

[Mark Meer]

They might not agree with your logic, or they might feel constrained by what they understand to be a conflicting rule or law.

(sigh). Agreed. We'll probably have to maintain several versions of firmware.

It seems to me that risk really should play a role here. If (as in this case) we are talking low-risk, where there are no possible hazards from mistaking the indicators, strict translation requirements seem to enter the realm of marketing/language-protection, and not safety/efficacy.

Putting the tight world of regulation aside for a moment, can we agree that from a practical standpoint, whether your on/off is indicated by "ON/OFF", "1/0", "Check-mark/X-mark", or "happyface/sadface", it really doesn't make much difference - assuming backed up by risk-analysis. In any case, the instruction are clear (i.e. "to turn on, select (insert chosen symbol above)"), and these instructions are available in any language.

 

[MIREGMGR]

(...) strict translation requirements seem to enter the realm of marketing/language-protection, and not safety/efficacy.

Yes, that's clear, I think.

can we agree that from a practical standpoint, whether your on/off is indicated by "ON/OFF", "1/0", "Check-mark/X-mark", or "happyface/sadface", it really doesn't make much difference - assuming backed up by risk-analysis. In any case, the instruction are clear (i.e. "to turn on, select (insert chosen symbol above)"), and these instructions are available in any language.

Yes, you can define any symbol you want in a language-specific IFU. You just (I think) can't define a text string as a symbol.