
On-the-fly translation is one of four possible approaches to translation.
If you are considering translating via on-the-fly worklists, note that there are certain constraints to be taken into consideration when compared with translation via standard worklists after setting up the translation environment fully. The table below outlines the main differences between these two approaches to translation:
| Standard Worklist After Full System Setup | On-The-Fly Worklist With No System Setup |
|---|---|
| You have to set up the translation environment fully before you can generate a standard worklist. | You do not have to set up the translation environment before creating an on-the-fly worklist. |
| There is a lot to learn when setting up the translation environment fully. | Learning how to create an on-the-fly worklist is relatively quick and easy. You only have to learn some of the basic concepts of online translation such as object lists, proposal pools, and working in the short and long text editors. |
| If your translation project is under high time pressure, you may not have time to set up the translation environment fully. | On-the-fly translation can help you translate more quickly than via a standard worklist, but only if you need to translate a relatively small number of objects into very few target languages. |
| You can take full advantage of various cost-effective and time-saving functions, such as DEMS Objects for Productivity Improvement (Top Texts), DEMS Objects for Quality Assurance, and Automatic Distribution. | You cannot take full advantage of such cost-effective and time-saving functions when translating via an on-the-fly worklist. |
| You can distribute the translation workload by assigning collections to translators in transaction LXE_MASTER. | You cannot distribute the translation workload by assigning collections to translators in transaction LXE_MASTER. |
| Translation statistics enable individual translators to monitor their own translation workloads, and translation coordinators to monitor the overall translation workload for one or more target languages. | There are no translation statistics. |
| You reserve objects in a standard worklist. Once you have reserved objects in your standard worklist, for example for target language Thai, no other translator into this target language can reserve the objects in his or her standard worklist. | You load objects in an on-the-fly worklist. This does not prevent another translator into the same target language from loading the same objects in another on-the-fly worklist, or from reserving the same objects in his or her standard worklist. |
| Standard worklists can deal with huge translation volumes into a very high number of target languages. | On-the-fly worklists are not suitable for large translation volumes. It can take a very long time to create an on-the-fly worklist and if you try to create one for too many objects, the program may exceed the maximum permitted runtime and terminate. |
Standard worklists and on-the-fly worklists are not mutually exclusive. In certain situations, creating an on-the-fly worklist can be useful even though most translation work is still being done via a standard worklist.
For example, standard worklist 0001 contains all of the translation-relevant objects in the system, but a developer needs specific objects to be translated immediately, that is, several weeks before the translation deadline. The developer’s objects are in worklist 0001 already, but the translators have no way of identifying them. The developer can name the transport request that contains the objects requiring immediate translation.
To solve the problem, one translator creates an on-the-fly worklist via the transport request while all other translators continue to work via the standard worklist.