Evan makes many useful points. I am the only official maintainer of the system and I only signed on to be backup. I just haven't taken the time to figure everything out. What's wrong with the current system? - there are (rarely) messages that require moderation on the mailing list and that isn't being done in a timely fashion. We have volunteers but the board has not yet considered them (I asked the board to do so). - a particularly important example is that Alan hasn't been whitelisted for the announce list. I will endeavour to do this. I haven't yet invested the time to figure out how to do this. - Gitlab hosts our meeting minutes and agendas. The new board members, particularly Evan, don't seem to like that collaborative tool. (I'm comfortable enough with it; it seems easy to me.) - at some point we are going to have to update the system to a newer releases of software (mostly: debian, mailman). As usual, there will be unknown hazards there. (Scott has thought about this.) In my estimation, switching to a variety of other tools will cause at least as many problems as they will solve. The key issues are: - we need a way of attracting volunteers (and new members; I will pretend that that is a separate problem) - get volunteer time and effort to provide services that work for us. "Us" means maintainers and users. - get enough volunteers so the we can survive unexpected resignations. - help volunteers do the right thing: work towards positive goals. - select tools that provide good services, ones that the "users" are happy to use. - ease "users" into becoming comfortable and productive with the tools - the tools must have a path into the indefinite future. Forced migration is not fun. It seems that the scarce resource is competent admins. Someone with enough time can become competent, so the main scarcity is time. We do seem to burn out competent people. We all leave them to do their tasks until that happens. We really need a pipeline of people to step forward. The task are not all that hard but when a small number of folks take on all the load, we get problems because successors are not being trained. I personally think that the existing services are mostly fine but need more: - maintainer cycles (currently my fault) - opportunities for users to master I am not going to be an admin of a non-Linux solution. ================ Just because "we've always done it this way" is not a sufficient argument. I'm not engaged in Discord but maybe others are. That's why Evan's push to create Discord channels is worth a try. Especially since he did the work. ================ Most of the time I've spent as a volunteer has been attending board meetings. I am not a member of the board but am invited to attend. That shows how little time I've actually put into maintenance. I should have spent more; I should have fixed the whitelisting problem. It is a worrisome sign that the board meetings sometimes don't reach quorum. | From: Evan Leibovitch via talk <talk@gtalug.org> | To: GTALUG Talk <talk@gtalug.org> | Cc: Evan Leibovitch <evanleibovitch@gmail.com> | Date: Tue, 29 Nov 2022 16:36:05 -0500 | Subject: [GTALUG] At the GTALUG AGM: How we handle Internet services | | [Apologies if you received this message more than once] | | Hello everyone, | | By now you may have read or heard elsewhere about the announcement of the | Annual General Meeting (AGM) for the Greater Toronto Linux User Group | (GTALUG), to be held virtually <https://blue.lpi.org/b/eva-zjc-gjy-kgl> on | Tuesday December 13. | | Along with other AGM business, we will be discussing how GTALUG goes | forward with our online services. | | TL;DR: A number of the people who used to set up and run our online | services (mainly the website and mailing lists) have left the GTALUG Board, | and those who remain have neither the time to fix nor familiarity with the | tools currently running these services. | | As Google has provided to GTALUG its nonprofit services which include full | access to commercial Google Workplace (formerly G Suite), the Board has | been investigating migrating some of our online services there. | | While to some such a move would be controversial, the status quo is | unsustainable. So either we need new people to step forward to operate what | exists, or we need to move to a service -- provided free of charge -- that | can dramatically simplify operations. | | The Board wants to have an open discussion on the issue before deciding, | and the issue will be raised at the AGM. We look forward to hearing your | views. While responses to this email will certainly be read, the full | discussion will take place at the AGM. | | Evan Leibovitch, Toronto Canada | @evanleibovitch / @el56 |

On 12/12/2022 18.34, Erica Peterson via talk wrote:

Re: discord -- perhaps, in the interest of supporting open protocols, we could create a Matrix space on element.io? It is free.

This would be preferable. Discord creates a large moderation burden (I've had to moderate a channel during events, and it is no fun at all), and the whole content is hidden behind walled-garden links Stewart

Hi all, The discussion so far has been very helpful. Many thanks to all who have participated. I look forward to seeing you at tomorrow's meeting. BTW, anyone thinking that I am "pushing" for Google is mistaken. I very much want to keep us using FOSS tools and using what we preach. It is an embarrassment to the technology we are advancing if we cannot use it to serve ourselves. However it is an even greater embarrassment if what we are using -- more likely, the way we are using it -- can't run basic things such as a few fairly simple mailing lists and a website that isn't borked like what showed up on my browser today (see attached). To me Google is just a highly-available last resort. The conversation so far has identified some people offering to step forward to help right the current ship and render it fit for use. That is very welcomed. Once properly configured, things *should* be easy to maintain but getting to that point of the challenge, Our needs are modest; if anything I wonder if some of our problems stem from over-engineering our solutions. Whether I personally am comfortable or not with Git (or any other specific tool) is or not is besides the point. What matters is whether (a) the tools we use are suited to the task at hand, and whether (b) we have enough people who have both the familiarity and the available spare time to produce the end results that we need. Having people saying "we could do XYZ", with the assumption that someone *else* among us has the skill and cycles to do it, is not helpful to us at this point. I am very hopeful, based on what I have read so far, that we will be able to come up with a way forward that will accomplish what we need. Evan Leibovitch, Toronto Canada @evanleibovitch / @el56

On Mon, Dec 12, 2022 at 09:02:09PM -0500, Znoteer via talk wrote:

Hi (znoteer the Montreal lurker here),

On Mon, Dec 12, 2022 at 07:02:04PM -0500, Evan Leibovitch via talk wrote:

...

Hi all,

The discussion so far has been very helpful. Many thanks to all who have participated. I look forward to seeing you at tomorrow's meeting.

[snip]

...

The conversation so far has identified some people offering to step forward to help right the current ship and render it fit for use. That is very [snip] Unfortunately, I have choir practice every Tues. I can possibly tune into BBB for the meeting tomorrow, but, alas, late.

Looks like I was too late for the meeting. I was the only one in the room. Sorry. Znoteer

On Wed, Dec 14, 2022 at 01:45:37AM -0500, D. Hugh Redelmeier via talk wrote:

| From: Znoteer via talk <talk@gtalug.org>

| Looks like I was too late for the meeting. I was the only one in the room. Sorry.

Don't worry, we elected you President :-)

LOL. But as President,, I declare the election was rigged :) -- Znoteer znoteer@mailbox.org

On Wed, 30 Nov 2022 03:39:51 -0500 Evan Leibovitch via talk <talk@gtalug.org> wrote:

On Tue, Nov 29, 2022 at 7:37 PM Stewart C. Russell via talk <talk@gtalug.org> wrote:

...

On 27/11/2022 21.01, Evan Leibovitch via talk wrote: Along with other AGM business, we will be discussing how GTALUG goes forward with our online services. I'm glad we're going to have this discussion. Chris Browne's untimely passing two years ago showed how heavily GTALUG relied on the work of a very few volunteers. And we do have a lot of online services, including:

- the website; - the mailing lists and their various archives. Maintaining and moderating a mailing list is no trivial thing; - GTALUG Wiki — https://wiki.gtalug.org/start — which I don't think anyone's touched for over two years.

It might be worth re-evaluating what resources we need. Could we make do with a hosted WordPress instance and move the lists to groups.io? All ideas are welcomed.

to admin a vps or two with apache/nginx and mailman with a few thousand? subscribers is, imnsho, not a major thing and in terms of vps costs is maybe like 10 bucks a month (for two vps) if even two is really needed I have been on the mailing list for ten? years now and I do not remember any month where gtalug had any volume that exceeds what a five bucks per month vps can carry and a properly configured mailman (which is very mature and can be battle hardened) cannot take more than an hour or two per month to admin. (that said, depending on configuration it can be a full time job - but of course, I am basing the hour or two on hardened configuration with some additional scripts, to do stuff automagically - as I am very lazy)

Anything we host ourselves bears both admin resources and financial hosting cost. Right now we're using mailman and frankly, I see the bounce messages and it's almost impossible to keep track of. (Hint: I tried mailing this Sunday night but that bounced). I really don't like mailman anymore. There are better ways to filter spam.

if your own email to mailman is bouncing, this is probably not what you think it is :)

...

Is GTALUG Inc a qualifying non-profit?

Yes. We've already been vetted by TechSoup *and* approved by Google. In fact we've had access to the Google nonprofit package for almost a year and could move things over to it at any time. The discussion at hand is whether we want to start really using it.

What happens when Google decides to stop providing this free service?

...

Google had indeed been known to axe products and projects that didn't pan out as they'd hoped (I particularly mourn the loss of Google Cardboard and Google Reader). Google for Nonprofits <https://www.google.com/nonprofits/> is unlikely to be one of them; it's actually very popular. A great many NPOs -- some of which are household names -- depend on that service. And with Alphabet being as insanely profitable as it is, the company is unlikely to jettison its primary CSR <https://en.wikipedia.org/wiki/Corporate_social_responsibility> tool.

That said, I'm in no position to guarantee anything.

and this sentence is the crux of it. personally I do not trust google. very frequently the left hand and the right hand are independently arguing with the right pinky toe over something the left ear heard the mouth whisper while a strong wind was blowing. it would be far easier to pay 10 bucks and get three sysadmins to each donate an hour a month (and their scripts :) ) so, if you decide to continue having an emailing list and self hosted services I would gladly donate some of my time (at least four hours a month), providing that there is at least one (or two) other high/senior skill dev/sys/ops also?

Where does our data go? “If you are not paying for it, you're not the

...

customer; you're the product being sold.”*

In this case, it's not the case.

When you use basic personal GMail and Google Docs, you are most certainly giving up privacy in return for "free" services. But that's not what's on offer here. We have access to Google Workplace <https://workspace.google.com/> (formerly known as G Suite) which is the commercial service that businesses use as a subscription-based productivity suite. In return for companies paying for Google Workplace, the company does not scan your data and it's subject to both Canadian and European privacy laws (after all, the apps themselves are pretty much the same as the free ones, what Workplace customers are paying for is privacy and support). What GTALUG has is a no-cost commercial license for Google Workplace (as well as other stuff including $10K worth of Google search advertising per month). So we have access to the commercial non-snooped service at no cost.

- Evan