Wedge

Which mode are you in?

I'd like to know where that rumour is coming from, so I can put a stop to it before it starts :whistle:

(4 files, 2KB)

Revision: 2086
Author: arantor
Date: 02 May 2013 14:36:56
Message:
Yay for thematic commits![1]

! The changes to the manage boards area for the moderation category as well as the new ignorable boards option. Of course, the support for the rest of this is not yet finished but every little bit I can commit helps, right? (Class-System.php, ManageBoards.php, ManageBoards language file)
----
Modified : /trunk/Sources/Class-System.php
Modified : /trunk/Sources/ManageBoards.php
Modified : /trunk/Themes/default/languages/ManageBoards.english.php
Modified : /trunk/Themes/default/languages/ManageBoards.french.php

1.	I should have put that here rather than in the actual changelog >_<

I sympathise, I've had my own run-ins with her ;) But ad-hominem attacks do weaken your own position when trying to argue the point.For the NPO to claim a legal stance on it, I don't see how it can be anything else. As Norv was an NPO member, I would argue there is probably grounds for editorial stance in that context, but that's not the end of the story by a long chalk.And that's the problem. The code does NOT belong to the NPO. It never has. They were licensed for use to the NPO.

Do you *own* Windows on your computer? No, you do not. You own a licence to use it on your computer. Do you *own* Linux on your server? No, you do not. You own a licence to use it on your server.

Just as I don't own SMF. I own a licence to use its code. You don't own SMF. You own a licence to use its code.

SM does not own its code. It owns a licence to use that code. That is what the CLA grants, and that's all it grants, once you distil the wording. The DCO is pretty explicit in this point, actually.

I own the things I make myself. I own large parts of Wedge's code. Nao owns large parts of Wedge's code. We don't own the totality of Wedge's code, we have a licence to use it in the way we are using it. The licence even permits us to relicence it but only on the totality of the work, even if we go closed source, the reality is that we're only relicensing our changes on top of SMF as a distributed work also incorporating other projects in conjunction with their licensing terms.

OK, here's where it gets interesting, and complicated.

We have some code in Wedge that was not written by Nao or myself. The notifications subsystem, as well as miscellaneous components (like jQuery).

In the case of jQuery, even though it is bundled, even though it is distributed with Wedge, it is not technically part of Wedge and even though our editorial decision was to bundle it, it is not legally under our copyright, nor can we claim it is part of the package for this reason.

In the case of Dragooon's notifications system... that's trickier. He originally put in a Wedge header listing copyright to the project, but technically that's not correct. It's copyrighted to him, and licensed to us via BSD. We do not and cannot hold copyright on it except for the changes we've made to it, and we cannot technically even claim copyright on the totality of the work for this reason because no aspect of copyright *law* grants us that right.

Remember, the BSD doesn't even allow us to remove SMF's copyright per se. We are still required to acknowledge it which is why it is important that it is correct. Though as stated, I'm not sure SM holds the copyright. I still believe it is the SMF contributors as a totality who hold that copyright since no copyright assignation was given to SM, only a licence to use the code.

See, you call Norv delusional but I'm really not sure she is.[1] I'd contact IP lawyers myself for clarity but I can't seem to find any in the UK that have experience in dealing with software IP.

I am still not convinced that SM can truly claim editorial capacity over SMF's code. Nao and I can make that claim for Wedge because we're not just platform stewards so to speak, we do not act in a primarily editorial capacity.

Nuts and bolts: was Norv ever officially a member of the NPO? What about Spuds or emanuele?

See, this is where it gets really complicated. For SM NPO to be able to claim editorial stance, I don't see how it can unless people who are members of SM NPO are the ones making the editorial changes. While the lead dev is a position appointed by the project, nothing the project does is automatically or magically transferred to the NPO.

For the sake of simplicity, let us assume that every person signed the stock CLA to the NPO. That CLA grants the NPO the right to use contributions. Not the right to claim ownership of them in any fashion, but simply the right to use them without any ability to remove them. Essentially it is the same as me agreeing I can't remove my own posts.

For SM NPO to claim editorial stance, members of the NPO in a duly appointed position need to be the ones accepting new patches, and not accepting others. Not the project team, but members of the NPO. Now, I don't know exactly who is and who isn't in the NPO but as per emanuele's post, a significant proportion of the team are not NPO members.

Now it is entirely possible that there are people performing the editorial role at this time under the NPO umbrella.

However, when the project was relicensed to BSD, was Norv a member then? Because if she wasn't, I do not see how SM can actually claim an editorial copyright at all - because it doesn't hold any copyright, not even editorial stance. Contributions were not given to SM, but licensed to them, and the terms of that licence indicated firmly that copyright was not transferred.

If Norv was not a member, then the situation is simply that it is still copyrighted to SMF contributors, not SM NPO, and edited under licence. SM's copyright is at that point almost a matter of sophistry.

If Norv was a member, that changes things somewhat. It is still copyrighted to SMF contributors above all else, and contributions were licensed to SM, but the position of editorial stance is somewhat more tenable. Though the copyright clearly needs to indicate contributors and I'd even argue it should indicate the licensing situation (that contributions were licensed to SM). The entire thing's a clusterfuck, no argument.That does not agree with other comments I have had from other team members.Tim and dtm, yes, even I would completely agree that they were not unfairly treated for their actions.

Also don't forget TE who felt he could not contribute as a developer unless he became part of the NPO but was not able to do that due to other matters.

1.	Though it's an ad-hominem attack and I'd hope you were better able to construct an argument than that.

test test test test test test test test test test

That's the point, there IS filtering. Every condition you add is a filter, that's how it works.

The problem isn't that you have a filter. It's that you have a filter that is mostly useless for filtering.

On the bulk of forums, this 'active' variable is going to be 'yes' on pretty much every post. Which means MySQL won't bother to use the index and will just happily go through every row trying to find the ones that are active.

On the other hand, if you explicitly tell it to find the values that aren't yes, the index will be very useful because it's doing what it would do best: finding rows selectively.

Even if we make it a single 'status' field that covers for both approval and soft deletion and even separate out deletion by user vs deletion by moderator for display purposes, that still means we have to deal with: active, waiting approval, soft deleted by user, soft deleted by moderator - 4 values (6 if you count waiting-approval-but-soft-deleted possibilities)

But the index is still going to be massively unbalanced because of the number of cases there will be of 'active' vs all the rest, to the point where MySQL won't touch the index. But it will explicitly have to go through every row manually to find 'active' as opposed to not having to explicitly check these things.


The sad truth is, what most people think is how MySQL works is not really how it works at all. There are entire textbooks on how to optimise queries by dealing with indexes.

It's not really about classes changing on condition, it's more about inclusion for things like the admin/guest/member/mod modifiers, so you define it once in the master stylesheet but only show that to the user if they're actually in that group.

Saves bandwidth for everyone else if you hide admin-only stuff away, that sort of thing.