Hi Mario,
over the years I have witnessed your amazing effort and I am sure many here would not be able to carry on their work without PTBs team continued help and upgrades. As it happens I woudl like to donate, but the purchase tag of 150 is somewhat to steep for a user in Romania (where I recently relocated). Needles to say, given the arcane funding mechanisms here, it would come out of my own pocket. Is there a way to make a smaller donation?
Thanks,
Tudor
Hi Tudor,
for smaller contributions by “private people”, there is currently still a way to do so, also listed on our companion website for all commercial services https://psychtoolbox.net - also under the section “Are you a user - Become part of the community”. The pure “Community Membership” without any priority support. It only allows you to download of a membership sticker that you can put on your labs website, e.g., like Natalia proudly does on hers (https://neurovision.uni-graz.at/en/). It would also allow participation in a potential future survey among paying community members. If we’d ever make enough money from our users and could build up any meaningful pot of money to fund development of interesting features and enhancements, we’d let paying members have a voice in what enhancements to make, could e.g., be voting among different new features i have in mind, to prioritize, or could be proposals we didn’t get yet. As things stand so far, community funding is so poor that we don’t have the problem of wondering how to spend it.
The direct link to the Digistore shop for this ~30 Euro (incl. sales tax etc) is here:
EDIT: We no longer offer these memberships, as they were losing us money instead of earning us some, as explained below.
However, i recently proposed to my boss to remove this tier, it may or may not happen, as he has the say in all things business. The reasons are three-fold:
I don’t want labs to buy this. It doesn’t contribute enough money, unless every single lab using PTB would do so reliably and consistently. So far at least 99.9% of labs seem to prefer to free-ride, contrary to what the feedback from close to 2000 labs over the last couple of years suggested. Our current 150 Euro offering is directly based on what people proposed to us, ie. what they said they would gladly pay for, roughly at that prize point per year.
Also the feedback was that labs can’t buy this, it is too close to a donation, so individuals would have to pay out of their own pockets, like you want to do.
So far, I got the feedback from my boss that after one year, at most half a dozen individuals ever bought that 30 Euro item, so it cost way more money to set up, than it made so far, or is expected to ever make, as far as i can see. So the whole thing is mostly a pointless distraction which makes the websites offering more confusing to understand, and extremely short attention span seems to be a major problem with many of our users - at least the ones deciding
In the end, the number of published articles citing use of PTB for their work is still rising every year since over 16 years (maybe Covid will put a dent in this due to lab lockdowns for last year though), over 3000 per year with a rising rate of publications per year as well. Knowing from experience that only every 2-3rd work that uses PTB actually cites it, possibly even less, we can be somewhat confident that PTB is used in more than 10.000 meaningful research projects every year. And PTB’s technology is also used in parts in Python based toolkits like PsychoPy, let alone my background work contributing to various improvements to Linux and other open-source components used for neuroscience research, ie. working with upstream projects. So my work not only benefits actual users of Psychtoolbox. If all this work doesn’t justify more than 0.1% of all labs paying a modest 150 Euros/year for an offering that should be purchasable by most labs without violating regulations, then i don’t know where to go from here.
An hour of often only sparsely available fMRI scanner time costs a lot more than what we’d like to have to make sure people have a good shot at not botching up that hour due to problems with stimulus presentation or response collection. “fMRI scanner” can be replaced with many other items in the multi-thousand Euros price range.
We are preparing another user survey to try to find out what is wrong with the current offering, or how to improve it, or how to turn the whole business model upside down. But if things continue as they are, we will have to make drastic and potentially painful changes, because at some point I shouldn’t be the only one suffering and sacrificing all the time…
In the end, you as a private person shouldn’t be required to pay for PTB out of your own pocket. Your lab and project funds should do that. Ofc. funding situation can be quite disparate, but the idea was that those labs who can afford spending a bit of money would do so, so little labs or individuals can still benefit.
Btw., just looking at your LinkedIn profile, because i seem to remember you as a ‘power user’, i saw that you previously worked at NIH, and the project and lab descriptions didn’t sound to me like a place that is on the verge of bankruptcy or so massively underfunded that 150 Euros/year would be a problem? How was the attitude towards this then and there?
Ok. Have to get off a train and have the first “less work” weekend in over three months.
Thanks for your support,
-mario
I was indeed at the NIH for 11 years, and before that in a HHMI lab for another 11. I don’t seem to recall a “buy” option the last time we had to install PTB, which is probably around 2016. Honestly, as molecular biologists we only very rarely dable into visual function or multielectrode array recordings, on which occasion we used PTB, which we have dutifully cited, or made our collaborators do so.
If I understand what you are saying correctly, well-funded labs are saying 150 is “too cheap” or not set up properly as something you can order, like a matlab license?
On the other hand, it sounds like, at the number of yearly citations, it should be possible to convince a funding agency that your work has a tremendous impact on the scientific community, and get financed that way ?
And, the NIH being a government agency, I don’t even want to fathom what would be involved to “make a donation” from government funds.
Good luck with your work and keeping PTB alive.
No, in 2016 all that existed was my PayPal donate button, which was too rarely used by users and labs. VPixx donated some money, a very few labs did and then some breadcrumbs from a few individuals, that’s it. Psychtoolbox development and maintenance from early 2013 - late 2018 was almost completely paid out of my own pocket, using up most of my life and retirement savings while living very frugally and foolishly rejecting high paying job offers in the Linux and open-source software field outside neuroscience. Trying to save Psychtoolbox, based on the encouraging feedback by many users and labs that they can’t donate money PayPal style, but if i would provide any kind of billable commercial service with proper invoices that would pass minimum requirements to be able to pay from grant money, lab funds etc., they would more than happily pay because Psychtoolbox is absolutely critical to their work.
We also had multiple funding workshops at the yearly VSS conference in Florida with hundreds of participants, coming to the same conclusion.
And a big user survey over multiple years, with close to 1400 labs participating. The outcome of that was again that if I’d provide some service that is billable, with invoice, and provides enough service to not be just a donation, at a not that high price point, so it can be paid from grant money, lab funds etc. (instead of the private pocket of individuals), then they would buy that service to support us. And mostly abstain from actually causing us work, so it would act as a contribution just to fund general development and maintenance.
Specifically almost exactly 33% of all polled labs said they definitely would do so. 50% said it would depend on the type and conditions of the service, 10% said no.
We also polled how much a lab would want to pay per year per lab, and that’s how we ended with the 150 Euros per year for a community membership with priority support.
The community membership comes with priority support specifically so it isn’t just a donation, which we know labs can’t do. Only private individuals can, and most won’t because it comes out of their own pockets.
So our most conservative prediction based on all data and feedback was that we should have sold at least 400 community memberships with priority support. A realistic prediction would have been more on the order 800 - 1000, but if one would assume that the numbers from the polled labs would also translate to labs that didn’t participate in the poll, then even more.
In order to keep the price low enough to be affordable for most labs without much administrative hassle, we kept the price at 150 Euro, also based on the assumption that > 80% all labs would not make use of the priority support. The other assumption was that a lab using priority support would not use up more than 30 minutes of my work time on average.
We mostly survived the last few years on contract work and some contributions by Mathworks, saving up the money (mostly by myself working very hard for an absurdly
low “survival level” salary) so we could finally build and offer a service to the community and partners that leads to actual sustainable funding and a salary and resources for myself which will not lead into poverty if i continue working on PTB and the whole eco-system around it, ie. stuff that not only benefits PTB users greatly.
No, well-funded labs don’t say anything. Some labs buy it, most don’t.
The real numbers are quite different, compared to our assumptions, which were all based on feedback from close to 2000 labs via survey, VSS workshops, and private communication:
Instead of 400 - 1000 licenses, we sold only 90! Less than a quarter of the worst case prediction.
So far 58% of the bought memberships are unused, although this is a running value, as a lab buying a membership can use it anywhere within 12 month of purchase. All we know for certain right now is that 14% of sold memberships were actual contributions without request for priority support - the memberships that run out right now.
Of those 42% (as opposed to the predicted 10-20%) who used priority support, way more work time was spent handling those requests than the 30 minutes at which it would have made sense: A request took at least 79 minutes to handle, but sometimes i forgot to enable request specific time tracking for extended periods of time, so it works out to more like 116 minutes per request.
At the current level, this doesn’t work as a reasonable and sustainable funding source. Apart from paying its own costs, It pays for at most 1 month of work time for actual development and maintenance of PTB. And only for at most 1 month, because my current salary is absurdly and unsustainable low. At a fair and appropriate salary it would barely cover its own costs ie. the work time for priority support.
Why it doesn’t work, we don’t know. COVID related lab shutdowns certainly didn’t help and aren’t helping now either. One hypothesis is that we don’t reach the right people who make these decisions. Ofc. communication channels are limited to
This forum - where it is prominently announced to each new visitor as top post, until one actively dismisses it. And the post had almost 400 reads.
Our website, Github issue tracker: Mentioned prominently at multiple places, hard to overlook.
Psychtoolbox itself since the 3.0.17 release mentions it in text at each ‘OpenWindow’ call, after each download,update,install or in case of trouble. Also a big eye-catching welcome screen drawing attention to it for at least 10 seconds every work session, so one has to be very blind to overlook it. With download/update rates in the thousands per year, one would assume many people see it.
General mailing-lists. We made announcements on a few, but can’t spam them all the time.
Still thousands of recipients vs. only 90 labs acting on it is hard to explain away as “nobody knew about it, how could they?” - Maybe the people making buying decisions are too far removed from the groundwork of research to notice this? Maybe grad students, post-docs etc. actually working with PTB don’t tell their PI’s?
Or maybe PI’s etc. simply don’t care and hope the problems solve themselves magically, or are so busy that they put off the 5 minute task of getting a membership infinitely.
Or maybe something is seriously wrong with our offer? But then we asked labs to give feedback if that is the case and so far we didn’t get anything about something being seriously wrong, only some small proposals for tweaks, often not really actionable.
We heard a few opinions of people who care and told us their opinion, or who we asked, but those who care enough are usually the ones who already did the right thing, because they possess reasoning skills and the gift of long-term thinking, and can’t understand themselves why their colleagues would act so short-sighted and so patently against their own longterm needs.
That’s just kicking the can further down the road. Such funding, if one gets it, can be useful to pay for specific projects - just like contract work - or maybe to fill some temporary gaps. Denis Pelli and David Brainard once told me that acquiring funding for PTB was painful, difficult and drawn out, when they did it in the early years, not an experience they would want to repeat. And i don’t even have the connections or on-paper credentials, or experience to do it, neither do we have the time or the time to wait out funding agencies decisions.
We need something sustainable, with a fair salary for myself, that allows to have a life in dignity and some security for later retirement, not something time-limited to 1-3 years, with the kind of minimum-wage salaries or absurdly far below industry standards salaries a grant will usually pay. I’ve been paid minimum wage (or the equivalent of it if one accounts for the actual work hours) my whole working life, while being exploited by my former academic employers, except for the 5 years where i wasn’t paid anything and exploiting myself. While i still enjoy my actual work deeply most of the time, I also know that I can not afford to continue full-time work on PTB and related projects under the current work and pay conditions forever. People with even remotely my skill set, experience and connections wouldn’t either. They shake their heads in utter disbelief when they hear about my work hours or compensation in relationship to my skill set.
And ideally you don’t want to fund just myself, but at least another person, and have some money for technical resources, so one isn’t constrained for even most basic equipment to used 2nd hand parts that are all almost all donated and between 3 and 12 years old.
Ofc. the company which employs me and provides the business infrastructure around all this - and also takes the business and legal risk - also wants to see at least some small profits. It is part of a non-profit foundation for the public good, but that doesn’t mean it shouldn’t make any profit, or lose money, just that profit would be reinvested in projects and sustainability of the company, instead of being paid out to shareholders or making a few owners rich. And so far the “save PTB” project over three years in is probably more a drain on their resources than a winner. So far PTB hasn’t made anybody even marginally rich, except probably Mathworks to some measurable degree.
Our community memberships can provide the needed funding, at least reasonably sustainable, without a 1-3 year expiration date like grants, for a small team, and hardware/software and other operating resources.
Also that model means that the goals of the people who pay for PTB’s development and upkeep and of the people who do the development and support work, and the people who actually use it day to day, are well aligned. That is not always the case with contract work and other one-off sources of money.
Most other options we came up, and so far rejected, have way more potential for conflict of interest between those who pay and the end-users which actually use PTB. Or way higher costs or risk to the research community. Or they exclude not so well funded labs and individuals.
That’s why it is designed to not be a donation. Actually i think we have at least one NIH associated lab which bought priority support. I generally don’t know who is behind the purchases, unless people announce themselves under their real names here, because we wanted to keep the sales/contract management and “actually doing the work” separate, to simplify things like data protection. Until somebody announces their name to me as part of a priority support request, they are just a randomly generated string of letters and numbers.
The pure donation part is just that - currently 30 Euros including tax - membership bit i pointed you to. And that also doesn’t take on at all, as i mentioned before. Some people suggested turning PTB into a public charity / foundation, so people who donate can deduct their donations from their tax, also if they buy the - currently 150 Euros + tax - bit out of their own pocket. But ofc. that is based on their assumption that they are not the only ones who want to pay out of their own pockets, and this would make it a bit cheaper for them. Our numbers say that almost nobody does that. And labs can’t donate, so the charity/foundation would do almost nothing for us, while increasing the administration/accounting/compliance overhead further for a charity or similar accepted as tax free. Boards of directors, board elections, constraints on what sources of income one can pursue in the future, what projects are possible, etc. etc.
On our website we also have a link to a merchandise shop for PTB themed coffee mugs, bags etc. The logo is kind of boring. I’m told we sold a total of one coffee cup for something like 2.50 Euros net profit. I doubt we should put much extra work into that one either. Maybe i’ll buy one for myself at some point, it might become a valuable rarity at some point.
Thanks. And thanks for your support!
But over three years into the PTB business, it is only partially luck at this point. It is mostly about labs wanting to keep the most advanced and capable research tool for data collection alive and kicking, and at least as important for myself, the eco-system supporting itself and its alternatives. Or if they care so little about the quality and efficiency of their own work, that they can’t be bothered to contribute a small and reasonable amount of money and time towards it.
Or at least provide actionable feedback that would allow us to change our offerings to match well and be sustainable if there is somethings seriously wrong with the current ones. However, given that some labs have contributed, both without any short-term gain for themselves, or purely out of urgent need for support, our current model seems to be not impossible to use if the will is there.
This response was way longer and more ranty than intended, but maybe at least it can serve as a reference for future discussions, or as catch-all answer when i get redundant suggestions which we already tried or considered and ruled out for a reason.
Best,
-mario