CoTech Fund proposals - year 2

Hi Chris, it's Szczepan here, thank you for your constructive objection.

• Agreed, it would be great to set up a review date and methodology for evaluating whether the system is useful to CoTech. What would satisfy your expectations in this regard?

• VPS1 should suffice to run it, can't speak on behalf of WebArchitects whether the price is going to remain the same. The 2-hour deployment fee is one-off (which is symbolical anyway comparing to how much work has already gone into this). Calum has worked with Happy Dev and has first-hand insights into Startinblox, which could help us finally deploy it (I would like to reiterate that this has been discussed since, at least, November 2018). I agree it would be useful to have a community building around it. I'd argue it's easier to invite people to participate if there's something tangible to participate in. I am happy to contribute to engaging CoTech members with the deployed system.
  Alternatively, would you like to propose how community building can advance without the instance running? To me this thread is the end of that road.

• Last year we were close to deploying Startinblox as a part of CoTech infrastructure. I think it will take at least a few months to assess whether the system delivers on functionality as well as whether there's a reasonable uptake. This is why the proposal states one year. As you implied, we are not talking just software here. It would become a part of our ecosystem, addressing an unmet need at CoTech. Happy Dev suggested they could give us access to their instance, I was personally reluctant to suggest this because Startinblox/Hubl has been presented in a variety of ways to tech co-op community, not just CoTech. The solution appears operational, the question is whether we would benefit from using it. This would require CoTech having a complete freedom to use the platform, including extending it if we find it appropriate. Here we arrive to the chicken-egg.

Looking forward to continuing the discussion.

Thank you for taking the time to answer my questions Szczepan!

What would satisfy your expectations in this regard?

I think I'd like to have a small group (2-4?) of people named as responsible for the project who can report back. It sounds like that's already the case. And just some simple measures of success. I really don't know much about the system by maybe:

• 10 CoTech co-ops listed in the first 3 months

• 5 successful skill swaps

I'm completely open to suggestions, my concern is that we'll effectively commit to paying £500/year for the future because we won't want to turn it off even if only 1 or 2 people are using it.

The 2-hour deployment fee is one-off (which is symbolical anyway comparing to how much work has already gone into this)

Yes, I figured that might be the case (thanks for all your hard work Calum!). You didn't mention anything about time for ongoing maintainance, I think we should cost that in if you think it would be required.

Alternatively, would you like to propose how community building can advance without the instance running?

No, you've addressed that concern, thank you!

I think it will take at least a few months to assess whether the system delivers on functionality as well as whether there's a reasonable uptake. This is why the proposal states one year. As you implied, we are not talking just software here

You make a convincing argument for running the software ourselves and not relying on Happy Dev's instance (I'd definately be interested in us paying them to host it for us though if that was an option).

For comparison an Amazon (I know) EC2 t3.micro instance (2vCPU, 1GB RAM, 16GB SSD storage) would cost $55 for a whole year. I know it's not a fair comparison and I'm not saying that we shouldn't use co-op infrastructure once we've decided it works well for us, I'd just like us to consider if there are ways to evaluate the software in use (which I agree is the most important next step) at a lower cost.

Thanks again for the input - I/we certainly won't be a blocker here, but I'd like to hear what others think.

Hey folks, really pleased this is moving ahead, thanks @Animorph Co-op.

In terms of measures of success, I would probably go for a number of people, rather than a number of coops who have put their data on the platform. So maybe something like 'at least 50 people, across at least 10 coops in the first 3 months' would be better.

I'm very happy to be one of the group who are pushing for and measuring uptake - particularly getting the SPACE4 coops on board and Outlandish of course. I might not be much help with regards to assessing its technical successes though.

I also agree that we should include a cost for maintenance. Do we know how much that would be for the year? I also feel that £105 is fairly measly since Callum has spent a fair bit of time finding out about this. I'd personally feel more comfortable if we reimbursed at least a day or 2 though it's not a critical concern.

No comments on the hosting side of things.

Thanks for taking the time to write up the proposal. I feel as if I'm missing a bit of context and so I have a couple of clarifying questions.

1. Could you explain a bit more what problem we're trying to solve or what opportunity we're trying to seize with Hubl?

2. Can you explain a bit about what functionality Hubl has and how it would help solve that problem or help seize that opportunity?

I'm definitely not against the proposal in principle, but I'd like to understand more about it before making a decision. Thank you!

Thanks for asking those questions @James Mead (Go Free Range) plus one from me on both and I also echo that I don't have any critical concerns as yet just need a bit more info about the why and the what before agreeing to another piece of software!

In terms of the proposal, we do need to be able to measure it and have a set limit or review date before committing to future years so James's first question about the problem it is trying to solve is particularly important, I think.

Hi, it's Szczepan again, thank you for all your input! Apologies for not providing answers to some of your questions in the proposal. I linked to relevant sources in order to encourage people to explore the landscape for themselves. There is no consistent narrative, various people have been involved in the process at different times and the work towards developing an effective skill sharing solution stalled. From the gathering in November 2018, Happy Dev & Startin'blox have been proposing a positive way forward.

RE reasons: problem and opportunity

• CoTech does not have a system for sharing skills in the network, thought it is possible to advertise jobs on community forum. Our internal skill mapping process has never been finalised, I believe it is because there was never a functional output to that work beyond a spreadsheet (it was a good one though). Due to its forum nature, Discourse is not suitable for this purpose.

• The only skill sharing in the network occurs between people who already know each other, which, in my opinion, is not very inclusive. On our join page we enumerate benefits of joining the network, but most of them hold true for just a handful of co-ops. Of course, the more you contribute to the network the more you can benefit, but rolling out Startin'blox/Hubl would give all members more equal opportunities. Over time we would generate a ledger of work opportunities and co-ops involved, useful data to study.

• With Hubl we could streamline the process of sharing jobs (between co-ops & individuals within them) and chatting to each other real-time (many member co-ops boycott Slack due to its proprietary license) to share knowledge and memes. I personally see Discourse forum as a largely outreach platform, while Hubl could be our internal system.

• Hubl could also be very useful in co-ordinating efforts, for instance if we decide to run (online?) gathering later in the year.

RE functionalities

Here you can see the Startin'blox features developed & used by Happy Dev. Hubl, the front-end implementation of Startin'blox, has 3 features out of the box (as listed by Alex in the thread I linked to in the proposal):

1. Team chat

2. Profile directory

3. Job board

A big advantage of setting up our system is that we can modify and extend it. However, given that more functionalities are coming to Hubl in due course, there is no need for us to proactively develop them, we could contribute to the code base on specific issues though.

A powerful long-term benefit lies in federating our instance with Happy Dev's and with other networks considering deploying Startlin'blox/Hubl (like FACTTIC). There might also be instances of Hubl in other sectors (such as housing), which would give us access to new markets.

RE maintenance & evaluation

• In reference to above, I do not see a need for maintaining software beyond updating our instance (which should be automatable). I am unable to provide an answer on how often this will occur, perhaps Calum could step in to provide us with info on how much time it could take.

• I agree there is a need to set a timeframe and evaluation criteria for the system. Since Polly volunteered to contribute to promoting adoption and measuring the uptake, I would suggest sticking to her criteria: 'at least 50 people, across at least 10 coops in the first 3 months'. However, I would also add suggestion by Chris regarding successful skill/job swaps. And a more qualitative aspect too. So there are at least two core aspects to evaluation:

  • how many people are using it - 50 workers from 10 co-ops after first quarter since deployment.

  • how many members benefited from using it:

    • directly: skills/jobs circulated - 5 per quarter starting from Q2, so 15 over the first year.

    • indirectly: usefulness of real-time chat system and other functionalities - survey to be circulated amongst CoTech Hubl registered users in Q4. 75% members would need to express positive opinions.

  We would publish a summary of the pilot towards the end of Q4 on our community forum and then decide on Loomio whether to continue hosting Hubl.

  How's this evaluation sits with you? Please feel most welcome to adjust the parameters, add yours etc.

  I am happy to be a part of the group working towards integrating Hubl to strengthen CoTech's capacity for internal collaboration.

  Hope this helps, curious to hear more thoughts on the topic.

Thanks for all this additional information! My main concern is addressed by the evaluation criteria, as long as we have something to point to next year before deciding whether to re-fund I'm happy.

I would like some thoughts on the suggestion to use a cheaper / pay-by-minute VPS in the first instance. I'm a little bit sceptical that this app (and its database / mail server etc) will run on a machine with 1GB ram, and maybe this would be a way to find out what resources we need? Of course it might be too difficult to migrate the data etc to a WebArchitects VPS when we're sure, in which case that's fine.

Hello! It's Calum here, I'm not speaking on behalf of my co-op- I've requested access with a personal account :-)

I would add to Szczepan's reasons that I had in mind that the "Channels" can also be used to help organise the "Guilds" passed during the Newcastle gathering (https://community.coops.tech/c/cotech/skills).. these are focussed around sharing skills, e.g. in training, but also towards social goals e.g. https://community.coops.tech/t/political-tech-projects-group/1849

I'm working on Startin'Blox (and thus have a conflict of interest!). I asked my colleagues yesterday about maintenance costs and it was proposed that Hubl can cover it for now. Below is the full email from Alex

I'm delighted to see CoTech committing some pounds to see Hubl happening. That really means something in terms of willingness to make the connection happen between our organizations, and I have a massive smile on my face when I read the discussion on the link you shared. That's really cool. Thank you guys for that. 

That being said, I believe CoTech should not spend a dime on this for now. 
What I mean is that we are investing quite heavily to make the deployment of a Hubl instance a breeze, literally a one liner today, and tomorrow even a simple click. We are also developing an interface, for which we won a european grant, so that users can migrate their data from one instance to another. 

I'm not trying to lock you guys in Startin'blox's server, you're free to self host and maintain Hubl in your corner, or even to use our server temporarily and migrate at a later stage. But we are already hosting and maintaining a dozen of instances on a freemium model, and we are processing it heavily so that the additional cost for us tends towards zero. It might be smarter to keep your savings now by benefiting from our free hosting and maintenance offer, and maybe invest in the development of a feature you'd like to have or in a bugfix you find painful. 

Again, I'm not fighting against CoTech hosting its own a Hubl instance, I find it really cool. I'm mostly trying to maximize the added value per pound spent. I believe we'd be better off with CoTech paying zero hosting and maintenance for now, and maybe investing in some developments or bug fixes, than with CoTech trying to redo the work we've just completed over the past few months in terms of processing maintenance and upgrading the app on each new release. As Hubl matures and contributions pile up, it will become easier and cheaper to host and maintain it, but for now I'd rather delegate that to us.

That's how I see it. 
I hope I'm being constructive and clear. If a call might help, I'm available at any time. 
In any case, it is really good to have you guys on board, and I can't wait to see it live. 

Take good care. 
Thanks for your time and commitment, that's awesome! 

Hosting in this way would still allow us to push custom changes to the server and/or front-end application as we wish. I agree with Alex that it's simplest/most cost-effective for the overall ensemble

My only concern is that "getting our money's worth" might have been a strong motivation to engage with the platform :-)

I think we should discuss later during the governance call, but on the other points I'm particularly in favour of having a working group for community animation (with or without Hubl!). There are some people in the Startin'Blox team who are focussed on this and so I would propose that collaboration between the two groups would be ideal! Maybe we could pull in others from FACTICC/elsewhere too, or certainly I think international co-operation should be a long-term goal

I think the strategy to review and the targets proposed are sensible. It would be good to monitor the volume of communication within the channels as well. I think a longer term e.g. one year would be necessary, because the application and community will evolve over this time

That's really great news Calum thank you for investigating. If we can let them host it for us, and instead spend some money or your (or someone else appropriate) time for setting things up and potentially contributing back any changes that would be great I think.

This all sounds really good to me, thanks everyone for taking the time to refine/elaborate on the proposal.

I like @Animorph Co-op suggestions for additional metrics for measuring the uptake/success. In terms of this point "directly: skills/jobs circulated " I would maybe just add the "successfully circulated" because if lots of stuff gets circulated, but to the wrong people/in the wrong way then I guess we have a problem.

Maybe another measure of success would be something along the lines of "CoTech has contributed at least 1 bug fix for feature development to the Hubl/S-B codebase" though not sure if this should be a deal-breaker to us using again in the future - more just a measurement of value when it comes to international cooperation ;)

Polly