CoTech skills mapping session at the gathering


#1

Having just had a useful discussion with @harry, one of the points Harry was stressing was putting some kind of quantification on how good someone is at any particular skill, rather than just listing minor interests equally with expert specialisms. One of the things we could discuss, alongside the basic structure for a skills framework, would be some kind of peer rating system, which should work much better for us as co-operating co-ops than it would do for competitive capitalist companies. Of course, to do this absolutely needs us to have a shared vocabulary (or taxonomy, or dare I say it ontology) and one of the things I’d like to focus on is how to create and (flexibly) maintain that vocabulary, as a commons.

But I write this without further consultation with @jdaviescoates, so please Josef feel free to disagree or add your own slant on this. What is clear from discussion with Harry is that it’s not practical to require everyone to attend a short session. It would be helpful, but maybe not helpful enough, given that many co-ops aren’t going to be there. So we can focus on the principles, the framework, and the processes (including motivations) for getting co-op members to fill in something that could be a real help to potential clients.


CoTech Space4 gathering 29-30th November 2018
CoTech Website improvement session at the gathering
#2

That’s an interesting matter! We’ve had lots of discussions about that at Happy Dev. We’ve decided to start with a declarative approach, where we can list skills and set a level (with only a few ones in “expert” and a lot if you like in “interested in”). This is completed by the list of people with whom one has worked before, which is probably the most important as it enables you to call them and ask if they recommend them. Peer rating is an interesting idea, but we fear it can be quite destructive if not used appropriately. We thought we would move slowly into it.

Looking forward to discussing that during the hack!


#3

Hi all - just dropping a note as I’m keen to join this discussion during the hack. Did I hear mention @harry that we might change the terminology a little? Call it something other than skills map? Either way I’m particularly keen to highlight some of the ‘softer’ or analytical skills also alongside the technical skill sets.

Looking forward to seeing you later this week.


#4

I think there may be a couple of related hacks.

I believe

  • @polly is doing a “what would a freelance co-op that acted as a recruitment agency within Space4/CoTech look like”
  • @asimong and I will be looking at what a useful ontology/model would look like
  • @sylvain and the Happy Dev mob will be looking at how the information about people and skills could be decentralised but searchable
  • Various Founders and Coders teams will be hacking potential ‘skills mapping’ interfaces

#5

I think just rating yourself would be quite useful too - specifically allocating the time you’ve spent gaining experience to the specific experience that you gained.

E.g. on paper I’ve been “doing PHP” for about 12 years - longer than @matt for example. However I’m a much worse PHP developer because I’ve spent most of those 12 years doing other stuff like business analysis, while Matt’s done mostly PHP.

One thing I think we could really do with is a taxonomy/ontology of tech-related skills. StackExchange has an exhaustive list of tags (e.g. https://data.stackexchange.com/stackoverflow/query/926850/get-popular-tags - hats off to them for letting us query their database!) but it’s too many and not organised. Wikipedia/DBpedia have fairly exhaustive coverage (e.g. https://en.wikipedia.org/wiki/Spring_Framework) and also have some hierarchy through category pages (e.g. https://en.wikipedia.org/wiki/Category:Programming_languages) but the categories are kind of dodgy in Wikipedia, and things get complicated around topics such as project mangement.


#6

I think Mozilla’s framework for this is quite interesting: https://twitter.com/Gankro/status/1046438955439271936


#7

If that’s Mozilla’s we may have over-complicated ours!

image

It is multidisciplinary though, rather than just technical, and it aims to be a bit more objective (e.g. list behaviours that would evidence various levels).

The Mozilla one is good for self/peer assessment but is definitely prone to the Dunning-Kruger effect and similar. As anyone who’s encountered the perfect “total confidence + total incompetence” storm will know!

We’d be interested in people’s thoughts about how our own framework (above) could be improved. Happy to do a session on this if useful or interesting for people.


#8

These examples of skills frameworks are really useful to have up to look at, thanks @Chrislowis and @Harry. My contribution in that vein is the European e-Competence Framework, which I’ve studied and analysed quite a bit.

I’d just like to check my assumptions for this thread…

  1. a group of us will be mostly working on this whole area for most of the two days
  2. we aim to end up with a reasonable agreement on an initial prototype map or framework, that probably won’t be complete, but that can be trialled across CoTech partners
  3. we’ll begin with a session where we share what we bring to the table, in terms of things like the above, particularly where people have worked with frameworks
  4. to start off, I will offer to share thinking around some fundamental issues that I have become aware of while working in this field, to help people orientate and start with an initial common language

One useful thing may well be to outline some “dimensions” of knowledge/skill/competence/experience. This deals with e.g. @harry’s distinction between historical time that one has been working with something, and the total “number of hours” put in, mainly resulting in acquired expertise.

Straight after people have checked in and put their offerings “on the table” we could have an exercise I’ve thought out to build up a first prototype bottom up. In essence, this would mean starting from individual experience, and honing the way we describe our own skills by insisting on broadening our language to include the experience of others, using language that we negotiate until we reach a working agreement. Anyway, I guess it’s a case of agree the agenda when we start working together on Thursday.

Anything too far off? Thanks for any feedback.


#9

To elaborate a little, how useful people’s knowledge/skill/competence is will generally improve with time spent, but different people will learn at different rates, and learning will also depend on how intense the learning environment is. So doing a couple of days each year one isn’t going to progress very quickly. On the other hand, the classroom experience of all theory and no practice is also sub-optimal.


#10

I generally agree, except with point 2 (we aim to end up with a reasonable agreement…)

I think we need to end up with an initial prototype – however minimal/viable it is as a product.

If we spend two days agreeing what we want then we need to spend five days planning, 10 days designing, 30 days building and 10 days testing and releasing something, and we don’t have those resources available.

I dare say that’s what you meant, but I just wanted to clarify.


#11

:smile: depends what you count as agreement! My guess is that we don’t disagree here. I am looking for a initial prototype as well. We need only to agree on what is necessary to produce the prototype, not looking for the kind of “agreement” that takes for ever… :persevere:


#12

If you do decide to build something minimal (and I think that’s a great idea) don’t discount building it as part of the new version of the website we’re working on. It’d be a shame to write a lot of bespoke code if we don’t need to. We should have a chat when you have an idea of what you want to do.


#13

I’m imagining we might build a spreadsheet or something :slight_smile:

If we get more ambitious we’ll try to make it Jekyll compatible.


#14

A spreadsheet sounds great!


#15

Two things I’d like to add to the map, perhaps not for the MVP but I think would be desirable to have eventually:

  1. A way to link people together who like working together. At Code-Operative, one of the things we want to offer to freelancers is the chance to work in a team that you enjoy working in, because you’ve worked with them before. We see this as one way to offer a better service than a standard contracting agency, if you can hire a group that you know already work well as a team.
  2. A way for someone to indicate that they’re currently available for voluntary projects, and what kind of voluntary projects they are (or aren’t) interested in. I’d like to use this map to find devs interested in contributing to Wobbly, and I’d like to know who would prefer not to be asked.

Another (possibly over-complicated) thing is, at Open in July, I caught the tail-end of a workshop around mapping projects, and how they often fail. I was wondering if some kind of recursion might work, like how DNS servers provide addresses to clients. If the mapping is some kind of continual process following a request.

Excited to talk about this on Thursday!


#16

Started a google doc for this sessions:


#17

Great session everyone.

Who captured the next step tasks?


#18

Hi Polly, thanks for your enthusiasm and help in the session.

I’ve started a wiki page at https://wiki.coops.tech/wiki/CoTech_Hack_2018/Skills_mapping

I’ll be aiming to add to it soon, but meanwhile (and after) everyone there is welcome to add their bit – would be be OK if I harmonise / copy edit any contributions as seems to me appropriate?


#19

Hi everyone who was at the skills mapping hack, or was interested but couldn’t make it.

I’m looking to

  1. form a small working group to lead this work forward
  2. a volunteer from each CoTech co-op to co-ordinate the skills mapping of their members.

Any offers? I’m offering to keep track of things, but of course I can’t do everything by myself, and besides, I’d like to be working as what Richard Bartlett (Loomio, Enspiral) calls a “crew”!

Many thanks

Simon


#20

I’d be happy to take this on for Outlandish.

Also happy to form a working group if you’re up for co-ordinating tasks.

You say on the write-up that “HappyDev’s site didn’t give us the detail that we thought was relevant”. I was wondering whether you could remind me the details on this? I thought that it had a lot of functionality we were looking for, and the ability for us to add in more. I think it would be good to aim to have something more dynamic than a spreadsheet soon, and Starting Blocks seems like a sensible place to start.