Gábor Szabó - גאבור סבו

I'm in the Perl community, but outside of TPF. So, as an "outsider", TPF looks like a buddy organization where 4-5 people approve some grants for small programming projects (that's the idea I got initially). And it's simply an unknown organisation for a real outsider (count managers, CTOs in LAMP companies too). So, this post and your previous ones are important to discuss this and while your grant proposal has a high amount of $$ (compared to previous TPF approvals -- and I see that this is the main criticism) I think that it's a good proposal.

These are all good questions, but you're mixing together things that occur on a variety of levels in a nonprofit, and I think that's really confusing.

First, you have some questions about organizational structure. How should the board be elected? This is a question of the organization's bylaws. Personally, I favor a self-selecting board as a way of ensuring continuity and avoiding hijacking. Maybe TPF could incorporate some sort of community nomination process? The idea of term limits sounds good until you realize that just finding people willing to be on the board (and actually do work as part of the board) is pretty damn hard.

Second, you have questions about the organization's mission. This is something TPF's board needs to decide, although I hope they'll listen to the community.

There's lots of stuff TPF can do. Let's look at what they're already doing.

First, there's the grants for Perl development. Broadly speaking, these grants are about supporting the technical development of Perl and related technologies (Rakudo, Parrot, CPAN modules, etc.). It seems obvious that TPF should continue doing this.

Second, TPF helps make conferences and workshops possible by providing some useful infrastructure. Again, it seems obvious TPF should continue with this. I imagine there's lots of ways to improve this program, too. For example, TPF could maintain a central conference sponsorship database to help each event find good sponsorship prospects.

TPF also operates some community infrastructure, including wikis, perl.org sites, etc. It's not clear how much of this is done under the auspices of TPF, and how much is done simply by independent volunteers. For example, who runs Perlmonks? How about perldoc.org? Okay, that's you, but is it a TPF project? It'd be good to have this documented more publicly, and also to make it easier for more people to contribute. Again, TPF should keep doing this. We've already seen some grant requests in this area, and I'd hope TPF remains open to these requests in the future.

TPF does legal work to help Perl, such as creating version 2.0 of the Artistic License, and working on Perl trademarks in various countries.

So we can see that TPF already does work on technical development, community support & infrastructure, and legal issues.

You're also proposing that TPF expand into several areas. First, we have general Perl promotion. Should TPF encourage the use of Perl? Can TPF help companies make _better_ use of Perl? I'm all for this.

Next, there's the idea of having TPF act as a bridge between the corporate and FS/OSS Perl world. This is related to Perl promotion, but not quite the same thing. In this case, TPF would be providing Perl-related services for companies. This seems like it could be useful, but I think it needs a lot more research to figure out what companies want, or if they want anything at all.

That raises the question of whether TPF can do any better than what's already there. A good example is Perl jobs and hiring. What can TPF do that the existing Perl jobs site can't do? Maybe TPF could spend some money to improve the site, as one example.

Third, you get into operations. It's kind of weird to ask if TPF should seek financing. Any nonprofit that wants to spend money needs to bring money in. The question, then, is what does TPF want to spend money on? To figure that out, TPF needs to clearly define its mission, and then work on a draft budget that gets into details of how it will pursue that mission. Working on a budget will help TPF clarify what programs it wants to pursue.

Right now it's not clear exactly what programs TPF actually has or wants, so it's not clear how much income is needed. But the question is not "should TPF fundraise?" The question is "what does TPF plan to spend money on, and how can TPF raise that money?"

Of course, one question to consider when contemplating a specific program is whether or not it will be fundable. Some programs may encourage donations more than others. In other cases, like YAPC or workshops, the programs are essentially self-funding.

Hiring employees is also an operational question. It's hard to answer this without first clarifying the mission and programs TPF is pursuing. This is also a question for the board. It's perfectly valid to say that at this stage, TPF wants to stay entirely volunteer-run. If TPF wants to hire an employee, then defining the mission and programs will help figure out what role(s) TPF should hire.

@autarch, I appreciate your detailed answer and the separation of issues but one thing I really missed from your post. What would you like the answers to be to the questions?

For several items you wrote that TPF board needs to decide and that you hope they listen to the community. So polease tell us what would you like them to hear?

@Gabor: I think I did answer some of those questions. I think what TPF is doing so far is good, and it should continue doing it. I'd like to see the grants program expand to include bigger grants, and potentially become a program where it could actually employ people to hack on Perl full time for a period of time (6 months, a year, etc.)

I'm also all for promoting Perl, but like I said elsewhere, I think the first step there is to come up with a marketing plan, not to actually start marketing.

As far as being a bridge between the corporate world and the community, I'm not sure if there's any demand for this. I'm also not sure if TPF has the bandwidth or skills to do this yet. Personally, I think this sort of thing can wait.

On fundraising, I think I've already talked about this at some length elsewhere. The first step for successful fundraising is figuring out how to market TPF. TPF does not do a very good job of communicating what it already does, much less future plans. Until this is resolved, it's somewhat pointless to talk about things like advisory boards, fundraising for bigger grants, etc.

As far as employees go, I don't think TPF is ready for this yet either. Managing an employee is hard. My animal rights group has a board which is _much_ more active than TPF's and we still have lots of problems managing employees.

Basically, TPF's board needs to get more active and figure out a bunch of other organizational stuff (mission, budget, fundraising) before it can realistically consider hiring people.

Why is your link for your proposal linking back to this post?

@Alberto, because I left out the actual link from the href="".

Now fixed. Thanks.

As Dave Rolsky points out, a lot of what you are wondering about it part of how non-profits and not-for-profits operate in the United States. TPF doesn't really want to be a membership organization. That's even more of a tarpit than the current situation.

Ideally, The Perl Foundation would just be a bank account and a checkbook. That is, they would be an actual foundation. Everything else is outside TPF so projects have maximum flexibility to do what they need to do. TPF provides the way for people to give money that can be funneled to projects, and a way to give money to projects. Anything outside of those two things would not be part of TPF. Individual projects and ventures should take on their own promotion. Indeed, most of the good stuff you in the Perl community did not come from TPF (or Perl mongers, Inc. or The Perl Institute). People JFDI.

As it is now, people, unreasonably I think, expect that TPF should be in charge of Perl and should work on anything and everything related to Perl. As people reinforce this idea, other people think they need to get TPF support or permission to pursue new ideas and projects. TPF shouldn't be in any position to get in the way of anything, and TPF shouldn't waste it's limited human time doing the work that projects should do for themselves.

Additionally, people keep talking about "marketing", despite that there is no "market" because no one can tell you what TPF even provides. It controls no projects and has no say in Perl development. It has no customers. TPF provides nothing. However, many Perl projects provide interesting things, but they all have different interests, goals, and communities. A federated "market" is just fantasy.

Part of the fundraising effort has to be tied to explaining what TPF does. The more varied the things under TPFs umbrella, the harder it is to explain. The more anyone makes it seem like TPF does anything other than act like a foundation, the more donors expect to get out of it. If I were running TPF (and remember that I ran Perl mongers), the tag line would be "We don't make Perl, we make Perl better", and my elevator pitch is that "We make Perl better by providing a way that Perl's users can donate money to fund the projects that are important to them. We underwrite major events to bring people together so the events can focus on content without worrying about the finances. We connect the people using Perl with the people making Perl because we make it our business to know everybody." That's it. Everything else should be pushed out to the individual projects.

Here's an example of how I think things should work. It's just an example:

1. The Catalyst project, on it's own and without TPF, markets itself and attracts donors.
2. Catalyst arranges with donors to make targeted donations through TPF, which just manages the bank account.
3. TPF gets the check, deposits it, and pays it out to Catalyst minus some administrative fee (2-5% is typical).
4. Catalyst uses the money, thanks the donor, and tells the community about it.

Here's how it's actually working, because once anyone creates an organization, people want to dump responsibility onto it:

1. A lot of people are waiting around for TPF to promote Catalyst. Most of TPF people aren't part of the Catalyst project though.
2. No one is donating big money to Catalyst.
3. Everyone whines about TPF not supporting Catalyst and 15 other projects.

Some of you say something like "TPF provides the way for people to give money that can be funneled to projects, and a way to give money to projects.", but people have to know about TPF to know that they can give money to them.

Maybe TPF is better known in the US, but what's with Europe, Asia, Africa? According to the Perl Survey both Germany and UK are big Perl "markets". Do the companies in Germany an the UK know anything about TPF?

I've sent a proposal to TPF about "local representatives" a few years ago. These local guys could do "marketing" (call it whatever you want). They could talk to people at non-Perl events and tell them "there is TPF and they make Perl better" (in the way brian described). But these people should be in an "official mission".

I disagree with the opinion that TPF should not do "marketing". In my opinion TPF should talk to others (people and businesses). And there was a grant proposal from Richard Dice (http://news.perlfoundation.org/files/090422_rdice_tpf_proposal_blog.pdf) that also describes "Marketing plan for TPF & Perl" as a goal (page 6 - "TPF internal work" -> strategy).

I've talked to a community manager of one company and he asked me "how can we give something back to Perl and the Perl community? We get a lot out of Perl and its community" (I showed him, how booking.com gave something back ;-) but he is not in the position to decide that).

Marketing is important and it would be great if TPF coordinates all the work (by providing *professional* merchadising materials or sources for that). I got lots of positive feedback for the Perl booths at CeBit and LinuxTage Berlin and now I plan to go to some schools to show teachers and pupils Perl (the initial contact was made at CeBit). The marketing/events group is great but it should be more professional (in terms of having a (maybe small) budget, having better promotion materials,...).

What TPF should *not* do is to put pressure on P5P. They are doing a great job (thanks to you guys). But it is great that the granted the money for Dave to do all the work on the hard bugs.

Overall I agree strongly with brian d foy. I don't understand why there's such a push to pile responsibility, commitment, pressure, and overhead onto TPF. Whatever happened to Open Source Culture?

Back in my day it used to be that if you liked a project you used it, then you talked about it, then you hacked on it, then you moved on. You did that. The singular you.

What happened when people didn't care enough about a project to do any of those things? It died, that's what. What's the practical impact of a dead open source project? Zero.

It's a striking sign that The Perl Foundation is expected to pay for open source contributors. What has happened to us? Is Perl so unloved that we have to pay for its upkeep? Back in my day that sort of thing lead to a dormant project. And so I ask, what's so wrong with that?

I want the TPF to do less. Right now TPF is using money to demotivate the Perl Community! It's killing the Perl. TPF should have less money, less initiative, and less central authority. In short, we all need to relax and enjoy ourselves.

Lets reclaim our intrinsic motivation to work on open source, to volunteer, to organize and form community. Money is the enemy.

Here is a great video explaining how we're really motivated. It uses the Open Source Community as an example.

Gabor:

I think the best way to approach what the Perl Foundation -- or any non-profit organization largely sustained by the efforts of volunteers -- should do is to ask a series of questions:

1. What things must it absolutely do?


2. What things are highly desirable for it to do, if not absolutely necessary?


3. What things are desirable for it to do if it can motivate its volunteer base?

Here are my answers:

1. Consistent with the laws of the country in which it is incorporated -- in this case, the United States -- it absolutely must safeguard Perl's intellectual property: copyright, trademarks, licenses, etc. That's it.


2. It is highly desirable that it perform functions which facilitate the activities of its volunteer base, in particular by doing things which avert the need for volunteer groups to be constantly re-inventing the wheel. Example: By maintaining checking accounts, professional insurance, central web sites, etc., it enables different volunteer groups in different cities each year to take on the organizing of YAPC and other conferences.


3. It is desirable -- but not necessary -- for TPF to do things like giving grants for particular projects. If there are volunteer groups which can raise the funds for such grants, then we can give them should we see fit. But if we don't have volunteers with fundraising skills and contacts, then we cannot. If we have people who are enthusiastic about going to non-Perl tech conferences and talking up Perl, then we should support such people. But if our volunteers prefer to put their efforts elsewhere, then that's where our support ought to go.

Over the course of my life I have participated in many volunteer-based organizations and have studied (and attempted to practice) community organizing. To survive, volunteer-based organizations must not first ask themselves, "What are the all the wonderful things we could be doing?". Rather, they must first ask themselves, "Consistent with our goals, what is the level of activity which our volunteer base can sustain over the next three- to five-year period?" Most volunteer-based organizations don't ask that question and, consequently, they don't last more than a few years.

So my belief is that TPF, like any other volunteer-based non-profit organization, ought to figure out what it can minimally do on a sustainable basis over a multi-year period and do just that. Anything else is icing on the cake.

A few other observations:

• A volunteer-based organization changes character when it hires staff. That change may be for the better or for the worse -- but it is a change. The staff has a strong motivation to keep the organization in existence in order to maintain their own employment.


• Given what I expect are the financial resources available to TPF, I suspect that if we were to hire an executive director, most of those resources would go to paying the director's salary!


• At this stage, whether we are a membership-based organization or not is probably not very important. (The Parrot Foundation, by the way, has been chartered as a membership-based non-profit organization. Perhaps in a few years we in the Parrot project will be able to say whether that was a good thing or not.)


• TPF is a type of organization with little precedent in human history: While chartered in one country, it acts on behalf of a worldwide community. Some of its officers/board members neither live in nor are citizens of the country of incorporation. It has to maintain working relationships with Perl-focused non-profit organizations chartered in other countries. It should act cautiously and prudently in this because there's not a lot of prior art.


• At any given point in time, TPF may have people who are very good at interacting with the corporate world -- or it may not. When we have such people, let's figure out how they can put our best face forward to the corporate world. But when we don't, let's not commit ourselves to a lot of promotion to corporations.
  

Gabor, I heartily applaud your enthusiastic and excellent efforts to spread the word about Perl. I would like to encourage other people to volunteer along side you. (I really liked Perl::Staff at YAPC::NA!) But I think that before we expand the activities to which we commit TPF, we first have to make sure it carries out its core mission much better.

Thanks for your thought-provoking post!

Many thanks for this post, Gabor.

One quick kvetch with a comment:

@autarch: Re: "I'm also all for promoting Perl, but like I said elsewhere, I think the first step there is to come up with a marketing plan, not to actually start marketing."

Couldn't disagree more. I think the work that Gabor has been doing -- personally making efforts to talk about Perl at non-Perl events, and reaching out to Perl projects to get them involved -- is the most practical, grassroots, and effective way to promote Perl.

Effective Perl promotion == less talk, more action.

Phillip.

What TPF could do? First of all the main principle is that it should be complementary to what the volunteer community can do, it should not compete with volunteer hackers. So for example as far as grants for hacking go - I would rather see it squeezed, constrained to tasks that have a very broad reach (and this can be, a bit arbitrarily defined as works on the Perl core or as very high level abstractions like PSGI) and that are currently visibly understaffed. TPF should not decide what libraries are to be developed - there is a well functioning evolutionary mechanism that does this.

TPF could provide infrastructure for projects, wikis for example, or servers to deploy demo applications, or maybe graphical elements like logos, design templates etc. There are many free wiki and forum farms - but they always come with rather offensive advertising - so I think there is a need for Perl branded farms. There is a free hosting service for the Perl Mongers groups - but that is only static HTML - a wiki farm would be an improvement here I think.
Some rather more improbable ideas - TPF could hire people with skills needed for hacking projects but ones that are not very common among hackers - so for example graphic designers, user interface experts, mailing list moderators etc - with the idea that Perl projects could ask them for help with no fee required.

I can also see there a need for some facilitation of the communication between the Perl community and big corporations. This is a tricky thing because of the differences of how they both operate. From one side a corporation manager could think it is a lose of time to talk with someone who does not have any authority to decide the direction the community would take, from the other TPF or any other body cannot command the volunteer hackers.

I am tempted to agree that the best way to attack this problem would be to let you Gabor proceed with your project and carefully observe the outcomes.

I would agree that TPF should contain and manage all funds for the perl community. For me grant money would be better spent on marketing. What TPF really needs to do is create a simple marketing plan and follow it.

1) Create a Image. The onion is fine
2) Create a slogan. "Perl is ALIVE"
3) Create a reason. Every IT home needs a foundation, build your IT infrastructure on perl.
4) Create a business plan. Why does perl need money? To prove to the rest of the world that it is ALIVE.
5) Build a customer list. Talk to Oracle, Symantic, IBM, Computer Associates and get major companies to sign on.
5) Execute the plan. Now RUN ADS. Pick the magazines that managers/CEO's/CIO's read; Information Week, News Week, New York Times Business.

This is just like fishing. Use some money to catch bigger money. The money money TPF has the more ad's we can run. The more ad's the more non-perl people really start to believe in perl.

"Ideally, The Perl Foundation would just be a bank account and a checkbook... TPF provides the way for people to give money that can be funneled to projects, and a way to give money to projects. Anything outside of those two things would not be part of TPF."

@Brian, this may (or may not) be the ideal situation, however at least according to The Perl Foundation Steering Committee Charter page:

The Committee shall support the following specific tasks:
1. Conferences
2. Communication
3. Fundraising
4. Financial
5. Awards
6. Language
7. Community

If I read that right, TPF does take onto itself a lot more than just transferring funds between donors and grant recipients. If that's the case, I'm not sure a more open membership would be a bad idea.

That said, in response to Gabor's original post - @Gabor, personally, I don't expect any of the things you listed ("should TPF try to ..."). TPF is a private entity and is free to do what it wants (although I assume if you donated a large amount you would get listened to :)).
Furthermore, the more goals and tasks you set it, the more funding it will need. Where's that going to come from? As the JPA story you linked to shows, even the simple activity of being responsible for a yearly YAPC in Japan takes all they have.

It was pointed out on another blog that I went too far by suggesting the TPF's grant making is "killing Perl." Sorry about that. I forgot how sensitive we are about Perl being dead or not.

Many companies I found perl mostly for sysadmins or QA testing automation.
Perl must have good frameworks architecture build sites.
It would be good idea to have server where we run web app for competitions

Many companies I found perl mostly for sysadmins or QA testing automation.
Perl must have good frameworks architecture build sites.
It would be good idea to have server where we run web app for competitions

Should TPF be involved in actively promoting Perl?

Perhaps TPF could start by cleaning up the things that hinder the community's promotion of Perl.

A start:

- convince the maintainer of www.cpan.org to bring the style into this millennium

- fix the whois record for perl.com

Another thing that just surfaced - I think TPF should talk to companies that break Perl APIs. This needs to be done diplomatically - but they should know that the Perl community has some weight.