Plans for Code Delta?
 

Does LoneWolf have plans for making patches Deltas (code changed) instead of Full installs (all code is replaced every update)?

It seems like it would reduce bandwidth requirements by a huge amount. Pathfinder is 88.9MB, probably going to get over 100MB by the end of this year.

It would be nice to see Patches as actual Patches instead of a full re-install.

Could help with license re-assignments as well.

On the surface, this looks like it's a great idea. However, changing to deltas would entail a great deal of work and typically yield minimal benefit. :(

The reason is that any changes to the underlying mechanics files for the game system must be propagated up through all the packages that utilize those mechanics. This means that most new releases will result in everything being rebuilt to incorporate the lower-level changes, which in turn means everything still needs to be downloaded anew.

So even if we invested the time to enable deltas, we wouldn't be able to leverage them very often. Compared to many of the other features already on our todo list, this one is unfortunately near the bottom due to its limited benefits in the majority of scenarios.

@Chiefweasel: This mechanism is completely unrelated to licensing and license re-assignments, so there would be zero benefits regarding license re-assignments by supporting deltas.

The comment was referring to, if the whole product wasnt installed with each update, you wouldnt need to check if the license was good. Just check it once when the product is installed and from then on you're good.

But the product itself is *not* installed with each update of the Pathfinder data files. There are regular updates of the Pathfinder files (and other game systems) that are completely independent of Hero Lab itself being updated. So the two are completely decoupled, and there is already no need to update the license key file when installing new data files, so we're already doing exactly what you just described.

Yes, there are times when both are updated at the same time, but that's the exception and not the rule. And it's the large size of the Pathfinder files that's the driving factor behind using deltas - not the comparatively tiny size of the Hero Lab installer.

Hope this makes sense...

Well, then perhaps you might want to look into why Deltas are not beneficial and build from there. In a good code world, you only need to deploy what has changed, and data would be separated from actual code. Data would not need to know about other data, and higher level (Top-Level UI) code doesn't need to know about lower level code (Modules, Addons, etc.) If this is the case, you might want to build an API rather than a link from the higher level to the module/addon.

The big thing is that when a data file changes, you're saying that everything has to be redeployed. This shouldn't be the case. I would never expect Bracers of Armor +1 to change, same with Level 1 Cleric.

The thing is that this system is very inefficient if this is the case. Each deployment you're wasting (NewCodeSize-(NewCodeSize-PreviousCodeSize))*Members in bandwidth.

I wonder if I could do a top-level code delta for my work computer that synchronizes only the data files from my home computer so I don't have to redownload the entire data file every 1-5 weeks.