| *** AlisonChaiken has joined #trustable | 03:49 | |
| *** ctbruce has joined #trustable | 08:27 | |
| *** ctbruce has quit IRC | 08:28 | |
| *** pedroalvarez has quit IRC | 08:29 | |
| *** pedroalvarez has joined #trustable | 08:29 | |
| *** ctbruce has joined #trustable | 08:32 | |
| *** laurenceurhegyi has joined #trustable | 09:10 | |
| paulsherwood | laurenceurhegyi: board not found | 09:18 |
|---|---|---|
| paulsherwood | pls can you make it public | 09:18 |
| laurenceurhegyi | Ah, I thought it was already. I'll sort that now. | 09:23 |
| laurenceurhegyi | Done. | 09:24 |
| paulsherwood | laurenceurhegyi: thanks. is there any way to create an alias/shortname for a board on trello? | 09:25 |
| paulsherwood | also dog-feeding =? dogfooding :) | 09:26 |
| laurenceurhegyi | I was never sure on the wording for that :) | 09:28 |
| laurenceurhegyi | Re the short-name - the board is currently called 'Trustable - Phase 1' which we can easily amend. But I'm not sure on an alias/shortname | 09:30 |
| laurenceurhegyi | I'll have a look now. | 09:30 |
| * paulsherwood was hoping for trello.com/trustable or similar | 09:31 | |
| laurenceurhegyi | Yes, I'll see what can be done. | 09:33 |
| laurenceurhegyi | Hmm - that string after trello.com appears to be the board ID, and it doesn't look like we can amend it. | 09:51 |
| laurenceurhegyi | I've shortened the board name to just 'Trustable', and I'll keep digging / ask on the user forum, but have a feeling it won't be available on the free version. | 09:53 |
| paulsherwood | ok thanks | 09:59 |
| *** laurenceurhegyi has quit IRC | 12:17 | |
| * paulsherwood has spent some *hours* trying to recover the links in www.trustable.io :/ | 12:47 | |
| rjek | paulsherwood: What's the problem? | 12:50 |
| paulsherwood | rjek: unclear. locally everything is fine. when i deploy, pelican seems to conclude PAGES is empty, or something | 12:53 |
| paulsherwood | I've hacked around it for now | 12:54 |
| rjek | I'm in the office; do you want me to take a look? I've no idea about Pelican at all really but some fresh eyes might help | 12:54 |
| paulsherwood | rjek: well it's 'fixed' now by my hack. you're welcome to take a look though | 12:57 |
| * rjek will continue with what he's doing if you're content for now :) | 12:57 | |
| paulsherwood | ack | 12:58 |
| paulsherwood | ChrisPolin: https://github.com/google/wycheproof | 14:56 |
| paulsherwood | actually that was just a general link, not for ChrisPolin specifically | 14:57 |
| ChrisPolin | Will have a looksee. | 14:57 |
| paulsherwood | ChrisPolin: how are you getting on with the picture of the elephant? :) | 14:58 |
| ChrisPolin | One ear at a time. =P There is a UML-rendered version of the workflow on the wiki page now, but it is still a png as I'm still trying to suss out how to get it to point to a renderer as you suggested. | 14:59 |
| *** laurenceurhegyi has joined #trustable | 14:59 | |
| ChrisPolin | The UML is in the repo now, so it can be changed as you like, but it's not as straightforward as changing it on the wiki page yet. | 14:59 |
| persia | If you're doing UML, please don't "paint" with it. You've used a state diagram to represent things that aren't states. | 15:01 |
| ChrisPolin | ack | 15:02 |
| * persia tries to relearn enough UML to provide useful suggestions | 15:02 | |
| persia | I think you want a "Component Diagram" | 15:04 |
| ChrisPolin | I'll have a play with it now. | 15:05 |
| ChrisPolin | Fixed. | 15:15 |
| persia | Much more readable :) | 15:17 |
| persia | (extra points for defining interfaces, but that is presumably another interation) | 15:17 |
| ChrisPolin | Cool. Still missing the components you suggested yesterday, I'm getting a better picture of that on paper and then I'll add them in, but it's now easily edited. | 15:18 |
| persia | Sure. My review of the change from drawing->UML was based on the assumption that the content was still matching that from yesterday morning. | 15:19 |
| ChrisPolin | Pretty much. | 15:19 |
| persia | Although my desire to use a component diagram was informed by reading related to our discussions yesterday about folk who are having success with machine-processed UML. | 15:19 |
| persia | (apparently lots of folk "paint", which makes the machine-processing folk unhappy, because they cannot consume the UML) | 15:20 |
| ChrisPolin | Haha I see the problem. | 15:20 |
| persia | Hrm. Reading more and thinking more, I wonder if the current diagram is a bit of a component diagram and a bit of an activity diagram, attempting to show both structural and behavioural elements of the expected systems. | 15:35 |
| *** AlisonChaiken has quit IRC | 15:52 | |
| *** sambishop has quit IRC | 16:05 | |
| persia | From interest, how was the choice of plantUML made? Following the chain of inquiry from above, I encountered a lot of references to textUML, and to various explanations why one didn't want to use XMI or HUTN, but not much on why one might prefer plantUML to other solutions. | 16:06 |
| paulsherwood | as with most things here, plantuml can be considered a strawman | 16:18 |
| paulsherwood | if there's something obviously better, we'd love to hear about it | 16:18 |
| *** sambishop has joined #trustable | 16:19 | |
| paulsherwood | textuml seems to require eclipse, though | 16:20 |
| persia | From the limited view I encountered, I think I like plantUML, but was hoping to find some reasoned comparison between it and textUML. | 16:20 |
| persia | Err, well, it's more complicated than that. plantUML has plugins for a wider range of display platforms and text editors. textUML seems to have a wider number of code generators (both implementation code and template code). | 16:21 |
| persia | For both representations, the majority of useful tooling seems to be integrated with eclipse, although there are ways to not use Eclipse. | 16:22 |
| persia | But my UML experience is dated enough that I don't have a good grasp as to which provides a better DSL for writing modern UML. | 16:22 |
| persia | (although, having read the HUTN standard, I can understand why nobody adopted it) | 16:23 |
| *** ctbruce has quit IRC | 16:25 | |
| paulsherwood | ChrisPolin: i think your 'Trustable System Structure' should be something more like 'Trustable Software Process' or 'Trustable Software Workflow'? | 16:34 |
| paulsherwood | i appreciate you're trying to describe infra for it also, but still, mentioning 'system' seems to lead to potential confusion - you're *not* describing a target system | 16:36 |
| persia | That confirms that it ought be an activity diagram, rather than a component diagram. | 16:38 |
| persia | Apologies for the misleading guidance previously. | 16:38 |
| *** sambishop has quit IRC | 16:43 | |
| *** sambishop has joined #trustable | 16:43 | |
| *** laurenceurhegyi has quit IRC | 17:11 | |
| persia | I find that the "Page History" links on the trustable wiki all return 404s. Are the page changes stored somewhere else for review? | 17:13 |
| jmacs | Same here | 17:14 |
| jmacs | You could clone https://gitlab.com/trustable/overview.wiki/ and view the revisions with git | 17:15 |
| ChrisPolin | Apologies, I've been scrolled up on the channel to reread yesterday's conversation and hadn't noticed the new posts. | 17:22 |
| ChrisPolin | paulsherwood, I agree that 'workflow' is a better description. | 17:24 |
| ChrisPolin | I'm getting the page history ok? | 17:25 |
| jmacs | It only appears to work if you're signed in. I suspect it's an error. | 17:27 |
| persia | And git seems to be a sensible workaround | 17:29 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!