thruflo

> I think there is great potential to build a distributed offline enabled database targeting mobile and PWAs with it. Something like PouchDB but with automatic conflict resolution.

Take a look at https://concordant.io

Yak driven development (YDD).

If you want to make sure you're doing something sufficiently differentiated, find an area where you still need to shave a few yaks.

Sounds like an opportunity for a seedlegals.com style sass.

There’s tipping and sponsorship infra, but is there a service to plug an OS project into corporate-friendly licensing and support invoicing?

I have an entry level 2019 Skoda Citigo. It is just a car, zero driver assist / digital gimmicks. The only digital features are a Bluetooth stereo and a CarPlay screen — but even they have physical buttons.

Thanks for the feedback. I get the same error. It’s a LiveView page and seems some bitrot has set in. Will fix.

Edit: fixed.

Congrats on the launch :)

The pricing incentives sound smart.

Do you see the ability to replicate more complex architecture as a differentiator? For example, one of your homepage quotes mentions "running a Kubernetes stack inside of a Layerfile". Can you elaborate more on how Layerfiles enable this?

An omni-directional "Poynting 4G-XPOL-A0001": https://www.amazon.co.uk/dp/B00C1DGFPS

They make lots of different types: https://poynting.tech/antenna-category/antennas/

Yup, definitely a head scratcher! So far I've been able to thread through an open window but when the weather turns I'll need to drill a hole somewhere at some point :)

The UN did a major study on the economics of ecosystem biodiversity, ie the value of nature and how to price it into the economy: http://www.teebweb.org

Hazy | Elixir Phoenix Developer | REMOTE | FT or PT CONTRACT | https://hazy.com

Hazy is a synthetic data company. You can get a sense of what we do from e.g.: https://hazy.com/blog/2020/03/23/synthetic-scooter-journeys https://hazy.com/blog/2020/04/03/generative-adversarial-netw...

We're looking for a hit-the-ground-running Phoenix developer to help us create a new web application. I'm the CTO, hit me up on my username at hazy.com

I’m a long time Python dev converted to Elixir (for ~3 years now, mainly web dev and data pipelines).

The tooling and developer experience in Elixir is first class and the ecosystem is highly mature.

Money makes money.

Say they raise enough to earn interest on their capital.

Say they make 1% interest per year.

Storing a GB costs $10. 1% of that is 10c.

How much does it cost to store a GB for a year?

Amazon charge ~1c/m for infrequent access and 0.4c/m for Glacier.

So somewhere between 12c and 5c a year.

https://powauth.com is a great Phoenix auth framework for API and HTML based apps.

It ships with gettext by default for i18n.

Yes, but starting from the position of “entirely artificial data”.

There’s only one “may” there but yes, it masks “potentially losing crucial information”. The post I linked to is pretty clear on that.

I totally agree that the business need is the driver and that people miss the imperatives if they look at it purely from a technical or mathematical lens.

In a sense, synthetic data is the least bad actually viable solution. The democracy of privacy / data agility :p

The idea that the synthetic generator model must have solved the actual modelling problem is a attractive idea that doesn’t correspond to what people want data for: they want to eyeball it, see what’s in it, test some algorithms, figure out how they might approach the problem. You can do that very well on realistic synthetic data (much better than any other privacy tech) even if the synthetic data has lost some utility through its statistical approximations.

The idea that training on synthetic data is a “charade” misunderstands the usefulness of having realistic, “drop in compatible” data that works with your existing code or models.

The ideas of training models as a service and also of working directly with synthetic data generators to “extract“ from them are great but incompatible with (a) the complexity of real world DS workflows in regulated industries and (b) data scientists current code / workflows / techniques.

If I’m a bank, I’m not going to give you my fraud rules and you’re not going to solve my problems with xgboost. Access to models is just as locked down as access to data.

This is why it’s useful having an intermediary. Like ... a generative model that you can train where the data is and then copy over to where the modelling is happening.

Yup. We focus on financial services, so time series data is a key requirement / capability.

There are nuances — like static vs dynamic distributions and whether the downstream model is actually just aggregating anyway.

I’m my username at hazy.com if you’d like to connect.

Mathematically valid but misses the business problem :)

Yup, people tend to confuse concepts and refer to synthetic data as anonymised data. They are very different things.

Anonymised data or redacted data are transformations of a data set that _hopes_ not to leak too much PII / sensitive data. People don’t use ML to anonymise but they do use ML to classify as a first step before splatting or generalising.

In that case, its absolutely right that the ML classifier not being 100% results in PII leaking.

This is a key reason why anonymisation and redaction are widely seen as problematic and are being replaced by synthetic data and, maybe in future, homomorphic encryption.

Gartner estimate that by 2022, 40% of AI/ML models will be trained on synthetic data.

Just because it may have some drop in utility today from real data, there are all sorts of scenarios where that’s outweighed by speed and ease of working with fake data rather that tied up in red tape real data.

See for example: https://hazy.com/blog/2019/12/09/data-science-on-test-data

Also, synthetic data can be augmented, rebalanced etc. Which is why in future it may well perform better than real data for data science work.

For example, think about this person does not exist and then apply that to business data.

Disclosure: Hazy cofounder. We’ve been doing smart synthetic data for a few years now.

That’s not how it works. Synthetic data is entirely artificial rather than transformed, so you’re not worried about “missing some PII”.

See for example the videos on https://hazy.com/product

Disclosure: Hazy cofounder.

I made one of these once: https://github.com/thruflo/ntorque

As other comments have pointed out, HTTP is not a great idea because timeouts, etc.

The one subtle benefit that I can relay is that by using your main database as the storage layer, you can enqueue tasks within a transaction, as per: https://github.com/thruflo/ntorque/blob/master/src/ntorque/c...

This reminds me of the recent YC post on the Token Effect [1], where they were arguing that tokens provide a way for decentralised protocols to capture the value of the network.

It's dead right that open source provides tremendous value for companies and that open source projects tend to capture only a fraction of the value they create.

This project looks like a much simpler response to the problem: if a handful of companies signup to support the platform then the platform will make it easy for other companies to support the development of the software their businesses rely on.

[1] https://blog.ycombinator.com/the-token-effect/

That's a pretty good idea, actually. Especially if you're working in a big company. It might take a while to get things off the ground, but selecting a CMS should not be JUST a technical problem; it should be an organizational "problem".

Where does it say proof of concept? The repo describes itself as a first implementation of the model described in the paper.

Opendesk | London or REMOTE | Lead Developer, Front-end Developer, Full-stack / Systems Developer

https://www.opendesk.cc is an open making platform based around workspace furniture. What's different about our furniture is that it's designed to be distributed as a digital file, downloaded and made locally anywhere in the world. We have a passionate early adopter community, a global network of professional makers and customers like Nike, Etsy, Arup and Digital Ocean.

We're mainly a London-based team of ~20 people working across web design & development, product design, CAD engineering and digital manufacturing. We're looking for a:

* Lead Developer * Front-end Developer * Full-stack / Systems Developer

## Lead Developer -- London, £40-65k

You'll lead the development of our online systems. These include a marketing website, a fulfilment marketplace, a design publishing and distribution system and various online integrations.

You need to be full-stack unix, deep on web development with experience across multiple products / projects / stacks and languages. It's a practical, hands-on role leading the team, coding, code reviewing, writing specs, collaborating with designers, making system design and architectural decisions and generally using technology to get things done.

## Front-end Developer -- London, £20-35k

You'll develop front-end interfaces for our customer, maker and designer audiences. You need to be a strong Javascript, CSS and HTML developer able to work with Angular directives and Ruby-based CMS templates. You're comfortable refactoring to improving performance and maintainability.

## Full-stack / Systems Developer -- London or Remote, £20-40k

You're a full-stack polyglot developer happy to design and develop back-end systems, integrations and automated workflows. You'll inherit a medium sized Python web application (Pyramid, SQLAlchemy), Ruby content management system (Spontaneous.io) and a domain specific design publishing and distribution system (see https://opendesk.github.io/winnow/). Your job is to maintain and extend these systems and generally develop and improve our web applications, online systems and infrastructure.

*

We're looking for talent, drive and passion for our model of open making. All these roles have solution design and decision making responsibility and everything we do is an iterative collaboration between creative technologists.

If that sounds like you then please do get in touch. To be clear: the CTO and Front-end roles must be based in London (we're in Hackney) whilst the back-end roles can be in London or remote. We're ideally looking for you to start in December or January.

If you're interested, email dev-recruitment@opendesk.cc with information about you (CV or profile links), the role you're applying for and a short personalised cover note explaining exactly why you're interested in it.

> By 2050, we will need to feed more than 10 billion people, requiring a 70% increase in global food production.

http://www.farm2050.com/

> We see the potential for Precision Farming to lift crop yields 70% by 2050 and create a $240 billion market for farm tech

Goldman Sachs, July 16 (https://www.dropbox.com/s/0kbn7fmqcopmh2n/GSR_071416_172225....)

You were walking past student accommodation in the summer holidays.

Perhaps LassPass users might like to donate to show their gratitude -- after all, you just protected an awful lot of people's passwords.

Have you got a mechanism you could post here for them to do so?

Below the intro paragraph on the homepage are two links: "Learn more" and "jump to the technical docs".

Given your aversion to the technical docs, I'd imagine you might click "learn more". This scrolls the page down to another lead in para with links to "overview" and "beliefs and principles".

So sure, it's two clicks not one but are you seriously telling me the https://urbit.org/posts/overview/ (posted on HN the other day) isn't a well written explanation of what Urbit is / does / why you'd be interested in it?

[edit]: typo.