$50k fast funding program for low-cost science tools

Hi GOSH community,

@jcm80 and I are working with Experiment to provide $50,000USD (up to $10k per award) in funding for new low-cost science tool ideas. Applications are due by May 31, but they are considered as they come in, so it’s best to submit as early as possible. More info: Experiment



Hi Shannon,

I have a few questions after reading the website:

  • How new is new? Does it need to be a totally new idea for this grant, could it be an existing project under a certain maturity level? Could it be reducing the cost of an existing project?
  • If we set up a project aimed at a challenge grant like this. Do we budget a project to expect only income from the challenge, or could we think of a project that takes longer and looks to secure crowd funding as well
  • There is very little information on what the budget can be used for except that it can’t be used for overheads. There is also very little information on deliverables/oversight except saying there is no reporting for institutions. Is everything done on good faith?
  • We need to define a research question or hypothesis. This is very difficult for instrumentation. “Can we build a FTIR for under $1?” isn’t really a research question. Do they have advice for how we fill in this field?
  • If individuals are not in the US, Canada, UK, or Australia they need to fill in a form which asks “This can be a university, museum, or research institute”, however the website says “We can also send funds directly to an individual.” Does this mean that you have to be wither from one of these 4 countries, or in an institution?

Super cool!

True. I’ve been thinking about a cheap BRUV design that kids can build. BRUVs are camera traps for water animals, but only scientists know that. Science educators have never heard of them!!


If anyone is looking for a project, some people started GitHub - mik-schutte/SimpleScienceSetup: This open-source hardware project is designed aims to create a starter kit for composing experimental setups. These can in turn be utilized for all kinds of purposes (e.g. education, prototyping, low-cost alternative).

The idea is to replace some (expensive) building elements used in science labs with plastic made cheap counterparts, project started with @jerzeek in Delft.

The repo is still a bit empty in terms of contribution infos (sorry for that, my todo list got full these days), and we do not plan to go on with the project, but anyone who think it is a good idea that deserves development is welcome to take over :wink:

1 Like

Hi Julian!

Hopping in as I’m seeing this before Shannon due to UK timezone :slight_smile:

All of the above are eligible, if the outcome is a low-cost tool for science.

In principle the whole amount could be provided by the challenge but that will be the less common outcome, particularly for those looking for the higher end of the $10k cap, because then we could only support five projects. We’re anticipating that at least part of the budget for most projects would be crowd-funded via the platform, and the challenge grant is helping projects along to reach their funding goal.

On the research question front, I’m not sure where that appears on the front end so I’m not sure what to suggest but “Can I build an x for $x?” would be OK by us.

Experiment.com would be the best people to answer these, we’re selecting projects for funding but we haven’t been involved in setting up the challenge programme. This seems to be the main way to contact them:
Feedback | Experiment

I hope that helps!



Hello from France !
I think we are going to apply with @sgnbraillerap to push the #braillerap project further.
The braillerap is a great open hardware braille printer that is able not only to emboss braille at very low cost, but also able to emboss vectors (maps, animals forms, mathematics and geometry). We sat up an humanitary mission in Cameroon in 2022 to build 6 machines and empower people and communities to use and repair them with the national blind association in Cameroon.
Now the idea is to make the machine able to bel launched by a blind person alone and to translate texts into more than 200 braille langages variations.
So if you want to make your places and work accessible : support Braillerap and associate the blinds to your workshops and discoveries !

Github of the project : Machine GitHub - braillerap/BrailleRap: Open Source Braille embosser
Software : GitHub - braillerap/AccessBrailleRAP: Braille transcription software for BrailleRAP. AccessBrailleRAP allow you to input text, translate the text into Braille and 'print' the Braille on a BrailleRAP. This software is compatible with NVDA screen reader.
Documentation in english : BrailleRAP

All the best !


Hi @hugobiwan
great idea we need to improve AccessBrailleRAP https://github.com/braillerap/AccessBrailleRAP software and start DesktopBrailleRAP to mix graphics and Braille giving BrailleRAP users an ergonomic tools to build efficient tactile material.

Hi All,

@hikinghack and I took the time to write a project for this grant and had it thrown back because developing instrumentation is not a research question:

I have to say I am fed up with this sort of hierarchy of scientific endeavour where the instrumentation we need is “not science”. I and have replied accordingly:

Beware anyone taking the time to write a grant for this which focuses on Open Science Hardware, unless we can change this purist attitude I don’t see their platform being an appropriate place for GOSH to affiliate with.

@jcm80 @shannond Sorry for going from 0 to 100 with my response. I’ve been driven out of my career by this sort of attitude, I never expected it to surface again here.


Hi Julian and all,

David from the Experiment Foundation here. I’m sorry the process was frustrating. Allow me to add more context.

First of all, I’m a fellow tool builder. I started OpenROV and other ocean tech projects, and I’m obsessed with making tool-building more respected and encouraged within the scientific establishment. (I just wrote a long essay arguing for exactly that! What should science cost? - by David Lang

I have had multiple discussions with the Experiment team about how to encourage and support more tool builders on the site. What we kept coming back to: when we’ve pushed these proejcts to be framed as a testable hypothesis, they result in much better projects.

On my end: I’ll help Jenny and Shannon add language to the grant application page with examples and an explanation. We can do a better job managing expectations.

I know it can be frustrating to get that response, but please know it’s coming from a place of wanting the project to be more successful. We’re here to help. Please feel free to reach out to me with any questions – david@experiment.com



Re: countries.

Our payment systems require that a bank account is in one of those countries, but the project can be anywhere (note: Experiment and the Experiment Foundation are subject to US laws around sanctioned countries).

Re: individuals.

The Experiment Foundation (separate from Experiment, and where the grant will come from) can fund both non-profits and individuals. We cannot fund for-profits.

Re: reporting.

The Experiment Foundation (again, where the grant comes from) requires reporting in the form of “Lab Notes” on the Experiment page. So the grant application is also the grant reporting platform.


I’m sorry to say this David, but your comment and the website’s approach is not helping any of these projects succeed.

I also fully adhere to Julian’s reply.

Any project about instrumentation has a “research” question: ¿can better data be acquired with better and more accesible tools? ¿can this technology be used for data acquisition?

This is not about expectations, it is about following through with what was advertised, and being consistent with your own beliefs.


The funds are meant to be used for that specific purpose. If not to the projects you rejected: to which kind of projects would you grant the funds to?

Please reconsider your stance.

PS: if you need a clearly stated scientific hypothesis-driven process for instrumentation, here is one :slight_smile:

  • Observation: Miniature piezoelectric devices have been used for linear actuation and nanopositioning.
  • Observation: Many-channel micropipettes require miniature actuators.
  • Hypothesis: Miniature piezoelectric devices can be used for many-channel micropipetting, and automation of large-scale scientific experiments.
  • Experiment: Test the hypothesis by building a many-channel micropipette using miniature piezoelectric actuators.

If this kind of project is not suitable, I want to know what kind of project would suit your and the challenge’s expectations simultaneously.


I had a link to an open source BRUV, but I forgot whose it was and where it was. I hope you make this happen!


I have to say David I couldn’t disagree more wholeheartedly. The best instruments are multifunctional and useful to a wide array of people. That should be baked into the design from the outset. If we concentrate on an instrument for one specific experiment it encourages corner cutting on any features that will not be used.

When I was last in Panama I did talk to a member of the Smithsonian institute who showed me the bees he captures and categorises in the laboratory. He showed me the microscope he uses. But it cannot be carried into the rainforest. We did try to use the OpenFlexure for this, but it was the wrong type of microscope.

Andy, one of the members of this project lives in Panama. He knows many of the researchers in the Smithsonian so he is able to get feedback from a vast number of users rather than one project. Taking a user centred approach to instrument development we are able to know what is needed. This allows us to design a microscope that is robust, reusable, durable, and appropriate for multiple uses. If we take a hypothesis based approach then I don’t need to consider a long-term robust-reusable instrument. The instrument is by definition second class and I am able (perhaps expected) to neglect any features not needed for this hypothesis to be tested. This is bad instrumentation design, this is a poor use of our communities resources.

Coming back to:

Let’s play the game of looking at some instrumentation papers I have worked on and thinking of the hypothesis that would make them “much better projects”.

These examples are facetious, but they highlight the point of what I believe were excellent and useful scientific instruments, where no one set up a hypothesis.

In my career as a scientist I never met this narrow, pedantic “everything must be hypothesis driven” definition of research since I left high school and entered research universities (as undergraduate, post-graduate, and then staff). Fundamentally your approach sees instrumentation as second class in science. You have bought in to a false notion of purity which is unhelpful, and relegates us to the sidelines.

What I would like to know is why you allowed this challenge to be added to your platform, when the challenge clearly asks for tools to be designed, without the same level of review that rejected this communities projects? I assume it might have something to do with the 13% of the money that your platform gets to keep?


Hi @hugobiwan , The project sounds really cool! To improve your chances of success, though, I might recommend figuring out some scientific applications that you can point to for the braille project

1 Like

I understand your perspective. I share your opinion on supporting more tools.

I’m slightly concerned that this forum is not the best place to have this discussion, as it seems to be getting a bit heated and missing the nuance. Would it be helpful to have a public zoom call to discuss? I’d be happy to host that next week.

I know the Experiment team very well and the GOSH community too (I met many of you in Panama). Everyone has similar goals in wanting to create more access. Let’s figure out a good way forward.


Hi David.

Happy to join a zoom call next week. Preferably within the UK work day (though I can’t do Tuesdays).

I am sorry I didn’t get to meet you in Panama. I had recently lost my job at the university after my department basically took the same approach as Experiment has in defining instrumentation as not real science. This discussion is heated because it is people’s livelihoods at stake, and many of us have been burned before. If your platform doesn’t see us as scientists, our community should move its money elsewhere.

I look forward to talking through a way forward, but it has to accept that instrumentation development can be valid research.


Here’s me in my rare diplomatic hat:

I understand platforms of any kind wanting thing to be in a particular format. That’s understandable if anything just to help go through lots of applications or whatever.

It seems a key problem is the format being asked for seems different from that which was advertised. I think @davidtlang and team recognized that and are apologetic about this misinformation.

Going forward though we have identified a key philosophical problem in this format that many in the open science hardware community are uncomfortable with, and have even suffered many real world consequences because of the perpetuation of certain ideas of how scientific work is done.

My proposed remedy:

If experiment.com let’s people apply with open science hardware tools as tools themselves, but in the format of the application requests an example experiment that could be carried out, i think most people would be ok with that. There could still be an example hypothesis described for testing, and we can even use that example project for field testing.

But framed as an EXAMPLE project, it would still allow researchers to propose highly useful innovative tools that have broad or unknown applications.

If experiment.com is ok with this kind of reframing, it might make it a more welcoming place for open hardware innovators.


Thanks Andrew.

Yes, an example is one route that would definitely work.

I also think we can improve Experiment to be more supportive of tools. I’m hoping to gather the productive points here and bring them to the Experiment team for a discussion about how to create better guidelines.

I think everyone agrees on wanting to support these types of projects, we just need to figure out how to do it.

For @julianstirling and anyone else who wants to discuss this, I’ll host a zoom call on Wednesday 10th at 10:30am EST to discuss. Send me an email – david@experiment.com --if you want to join. We can record it and share, too, if you want to send questions that you’d like me to cover.


Top diplomacy Andy. I look forward to the call. Partly because I want to see this rare diplomatic hat!!