Template talk:Cryptography navbox

Crypto templates

[edit]

So, I am going to try and sort out this mess. First of all to all who read this: These templates are the results of years of edits by the editors in the WikiProject Cryptography. And changes have been discussed and tested on the talkpage of that wikiproject.

Recently I did a major rework of them with a generic box at the botton and inserting the other boxes in the top of the generic box. So it would look better than just stacking 2-3 boxes on top of each other. This was tested and discussed properly on the crypto project talk page before deployed on the 300-400 pages that use them.

Then along came Ed_g2s and changed them without talking to any one. And without testing them properly so he broke them in several ways. But he also added some nice ideas. Unfortunately at the time I was going to bed and was to have a busy schedule for some time so I have not had the possibility to handle this until now. So at the time I just reverted them so the 300-400 pages would not be broken. --David Göthberg 01:29, 29 October 2006 (UTC)[reply]

Here follows a discussion cut and pasted from Ed's and David's talk pages:

Hi Ed g2s.

Thanks for giving me todays big laugh. I just saw that you edited the crypto navigation boxes in WikiProject Cryptography. You actually broke those templates in a whole bunch of ways. Too many for me to list right now since I am just about to go to bed. So for now I will just revert those templates so the 400 pages or so that use them will look alright again.

Tomorrow I will go over your suggested changes in detail and see which changes are nice to use and test them properly on my own sandbox pages before deploying them. Since I will revert them now, here are two screendumps so you can see how they looked in some articles: broke01.png broke02.png

Note that those templates are more complex than one might first think and they are used in several different ways on many different pages. Every single character in those templates are there for a reason. I really did spend a lot of time on testing them and reducing the code to the bare minimum and still making them look exactly as we wanted them AND making them function in all the usage cases we have for them.

I am off-course partly to blame since I have not yet documented them fully. And I should have added a warning on those pages "These templates use complex tricks and are used on about 400 pages. If you edit the code in these templates do tread lightly and test them on your own sandbox pages first." But I guess my excuse is that I made those templates three weeks ago and have had other matters to attend to since then.

For instance, you removed my workaround for the "pre-tag + nbsp" bug. (It's an old Mozilla bug that MS copied to be compatible so now both Firefox and IE has that bug.) I really have to write a comment about that workaround in those templates or people will remove that strange code again and again.

I noticed you changed some of the table "html-code" to CSS-code instead and some of those changes seemed nice, since that made the part of the templates where regular editors will add links look cleaner. I have actually spent the last few days reading up on CSS.

But why is every one so afraid of tables in tables that they/you prefer using DIV-tags in DIV-tags instead?

I'll come back to you later since I probably will have more questions.

--David Göthberg 15:26, 20 October 2006 (UTC)[reply]

Paraphrasing actual response: No, I didn't break them. ed g2stalk 15:54, 24 October 2006 (UTC)[reply]
I'm pretty sure you don't have the best set-up - if you would like to link to the various cases that broke, and explain the ways in which the template is used, then I can fix it. The first screenshot you posted just a problem with extra line breaks. The second problem looks like someone has included the template directly, which is wrong. I didn't change any pre tags or remove any nbsps, so I don't see how I can have broken any wrapping issues either. ed g2stalk 17:09, 20 October 2006 (UTC)[reply]
I have restored my edits, with the line-break fix mentioned, and checked the first 100 transclusions on FF and IE. As you can see on NESSIE, this is working fine. The second example you gave is a case of bad implementation, and not an error on my part. If there are any pages which are rendering incorrectly please inform me, and I can correct the code, but there is no need to revert all my edits. Thanks, ed g2stalk 17:19, 20 October 2006 (UTC)[reply]
I've also done an AWB search through the transclusions of {{crypto block}}, and it seems none of them include the template directly. The only page it found that did was one of your sandbox pages, so I don't know if this is a real problem... ed g2stalk 17:28, 20 October 2006 (UTC)[reply]
End of cut and paste from Ed's and David's talk pages.

Unifying VMAC and UMAC

[edit]

In the current crypto navbox the row on MACs looks like this:

 DAA   CBC-MAC   HMAC   OMAC/CMAC    PMAC   VMAC    UMAC   Poly1305-AES 

Indeed, the designs of OMAC and CMAC are quite similar and these MACs should be grouped together (with the slash). However, VMAC and UMAC are even closer (basically, it's the same design for a different word length), so I propose to group them together as well: "UMAC/VMAC". Ultimately, one should think about putting the UMAC and VMAC related contents onto a single wiki page. — Preceding unsigned comment added by 134.219.148.47 (talk) 23:58, 7 February 2014 (UTC)[reply]