Huge fan of Reticulum, fixes some of my biggest gripes with Meshtastic. Shame it hasn't got as much adoption yet. For those looking for Meshtastic-equivalent things in the Reticulum ecosystem:
Didn't got adoption because the code base is awful to work with and there is a trauma against bluetooth being used as a network path.
Plus: encryption is heavy when bandwidth is limited and over radio waves we aren't even permitted to encrypt data most of the times.
Please don't read my comment as bringing down the project. I'm a fan, used everything it was produced but ultimately is unusable for serious applications on the current state. I really tried hard to adopt it.
If the reticulum code is worse than the meshtastic one, then it is truly atrocious. Been trying to get a specific board to simply "sleep" its radio using meshtastic, and nobody seems to know WHY it doesnt do it. The code is horrible spaghetti with lots of ifdefs. And nobody seems to know why things are the way they are in the code re: power handling. ChatGPT wrote me a brute force method that works, but its ugly and I dont want to maintain patches.
But it is fairly easy to hack on. I have no idea how to debug things without USB serial connected, though.
Sorry, can't really compare because I've never had to suffer looking at meshtastic source code. Quite tempted at this point to just throw the python implementation of reticulum at Claude and see if a validated port to C++ is possible.
Maybe a bit offtopic and not LoRa, but I've been looking at ESP32 and they include an ESPMesh for the WiFi radio with a promise of about 500 to 1000 meters range from what I read. It isn't the same range as LoRa, but it is "larger" bandwidth and for the price of 3 dollars per unit seems promising on urban areas to connect people. I'm trying it out now.
What are those gripes? If I don't have anyone else who would use it, but would hang out in a public chat room, it didn't seem like reticulum was the right choice for that? You need destinations on things?
It seems like big cities get congested, on marginal systems the chances of only getting half the messages is very high. It really dosnt integrate with much else, the mqtt stuff seems unreliable.
It does seem like the RNode radios are a lot less mature but they seem to be aiming to be less of a toy.
Nomadnet it's really bad; it doesn't properly work with a 80x24 terminal and 16 colors.
Also, it uses tons of CPU on legacy machines. It needs some rework.
Not everyone it's a hipster with 256 or 32 bit colour terminals, shitty NerdFonts (nonstandards) and big displays.
And being written in Python3 makes it dog slow. Being rewritten in Go would get a few performance tweaks, (networking and GC there it's ideal), security and portability. But, please, no BubbleTea unless you can be sure it can work on a plain XTerm with 16 colors (I use Tango for readability, but 16 colors FFS). Keep 256 colours as an option.
The content seems really great, however the typesetting makes it quite hard to read:
* Code blocks on a subsequent page to the explanation, especially when there is enough space to show it (p18/19)
* Call-outs as above (p26/27)
* Single words broken by a page (p51/52)
* Footers spanning multiple pages (p61/62)
It sounds quite nitpickey but I find it really breaks the flow when I’m reading, and trying to comprehend a section requires scrolling back and forth between two pages.
Zod 4 supports converting a Zod schema to JSON-Schema (natively, this has always been possible with 3rd-party libs).
One key difference is preprocessing/refine. With Zod, you can provide a callback before running validation, which is super useful and can't be represented in JSON. This comes in handy more often than you'd think - e.g converting MM/DD/YYYY to DD/MM/YYYY before validating as date.
I’d imagine their capabilities mirror that of Mistral OCR [1]. Mistral outputs markdown, the image would have to be convertible to a reasonably useful markdown structure (charts, tables etc).
This highlights the biggest issue I've found with Mistral OCR. Many of the documents I upload are entirely classified as images, which means no OCR is being run.
Pretty much anything with a different colored background gets returned as (image)[image_001].
LLMs tend to be a hammer in search of a nail when it comes to documents that have imagery. We decided on CV models which results in a high 90s midpoint for the docs our customers care about. If you can afford to go with a cv pipeline, it can outperform all of the LLMs by some margin.
That works fine if the company itself stores the data, but becomes difficult to enforce when 3rd parties store the data. Imagine a company with an EU presence stores it's EU data in US, with a hypothetical cloud provider that doesn't have an EU presence.
The company would need to have a DPA with it's cloud provider. That cloud provider technically would also need a corresponding DPA with any 3rd parties that they themselves use, except without an EU presence that is hard to enforce.
In this case where there is one hop you could argue that it's the companies responsibility to ensure that their service providers are operating in compliance. Imagine the same scenario, but with one, two or more middlemen and the whole thing becomes an unenforceable mess of jurisdictions for the company to do meaningful due diligence on their service providers.
It's much easier for the EU to say EU data has to be stored in the EU, and know that any party touching the data is likely to be in compliance, and significantly easier to investigate if they are not.
There's also the Cloud act, which makes it illegal for US cloud providers to refuse data access requests from the US government.
As far as I understand, the EU is fine with you sending data to other countries, as long as those countries have the same standards for data protection. In the EU's opinion, the Cloud act, as well as the whole NSA situation, mean that the US doesn't fulfill this definition.
I just gave Martin a go. What I'm looking for is an AI PA that can:
- do some research on a given company/individual/website and give me a summary.
- preferably also identify a contact email.
- handle selecting a good time for meetings according to my availability and preferences.
- handle the communication with the other party.
- let me know when it is arranged, or if it's given up.
I signed up and gave it a UK phone number, and got a UK number back for texting Martin. I'm not sure why it has to be SMS when it could be an in-app chat. I was expecting to get a confirmation SMS or similar, but it just accepted it straight away. When I texted the number I was given (several times), it was delivered but there was no reply.
Martin sent me an email welcoming me. I replied asking it to set up a meeting for early next week with another email address. Martin replied saying it is unable to email people on my behalf, and suggested I set it up myself.
> Unfortunately, I am currently unable to send emails to other people on your behalf. However, you can easily send an email to ** to schedule the meeting for early next week in the afternoon.
I reminded Martin that there is an example on the website homepage of doing just that, and it replied saying it can indeed schedule meetings, and asked for the details again. I replied with the same details, and it confirmed the meeting was set up.
I checked my other email, and there was no message setting anything up. I told Martin that the other party needs to know about the email, and it replied with:
> Understood. I'll make sure to inform ** about the meeting details.
Still nothing received. Furthermore, I checked the app and I haven't even connected my calendar, so I'm surprised it didn't warn me or prompt me to do this when I asked for a meeting.
I gave up with that and decided to try something else. I forwarded Martin an email thread from a lead, which included a lot of back story on their organization, offering, and some areas that they think we could potentially collaborate on. I asked Martin to find out more about the company, and evaluate the options for collaboration.
This lead is in the AI space, with their primary product being a document digitisation solution to help surface and discover business documents.
Martin replied describing it as a "nearbound revenue platform to streamline revenue operations", with a key feature being "Automated lead scoring and distribution to prioritize high potential leads". As far as evaluating the collaboration opportunities, it instead gave me a list of collaboration features within the platform, none of which exist.
At the end, it linked to a blog post to their recent funding round. Except, the blog post was from a completely unrelated company with a similar name. Bear in mind that the originally forwarded email was from their business email account, and the body contained multiple links and references to their website.
I decided to try one more test, and asked it to do some research on my own business website and let me know what it finds out. It's been 20 minutes, and I haven't had a reply. I checked the app to see if there was any indication it's working on something for me, but nothing their either.
I love the idea of Martin, but I'll be canceling my trial - it just doesn't seem anywhere near ready yet - especially given I have to trust it to communicate on my behalf.
Thank you for all the feedback! Just wanted to hop on and address a few things.
- I totally resonate with your criteria for an AI PA - this is very much what we're working towards with our email integration. We had been focused on voice for a while, but recently started tackling all the email use cases. Really want to get these right for you!
- Sorry for the poor onboarding job - we should make it more clear that you have to sync your calendar before we send you an email inviting you to send and forward scheduling items to Martin.
- For sending emails to contacts, this is one of our upcoming integrations that we've been building for a while - but just not ready yet! We want to make it able to send/reply to emails and fully act on threads that you attach it to. This means issuing you a unique email address for "your Martin" and managing it's behavior on threads and memory of other contacts. It's a harder problem than we first anticipated, so we're working through it steadily! It should be ready in the next month or so. For now, the communications feature is just limited to texting contacts on your behalf.
- For "deep searches", it definitely isn't the greatest at digging into a topic or generating a thorough briefing for you right now. We're not sure how deep we'll go into this use case in the future, but we do plan on integrating with more specialized functions, like LinkedIn, Twitter, Maps, etc. which should make this a lot better.
Sorry again for the poor onboarding experience. I think we also got an email from you, so will reply there as well to ask for more feedback!
“It definitely isnt the greatest at digging into a topic”
I think this is the greatest drawback to using a tool like this. It does good with the low hanging fruit tasks like responding to a txt. But it does poorly at a loooong range of long tail tasks like conpany research, or summarizing a topic or gathering the latest headlines for a topic, or finding leads or alerting me when someone mentions my brand on Twitter or when someone links to my blog post.
And when soneone tries to do any of those things and gets a bad user experience, theyll give up on your tool
That is what valuable customer feedback looks like. I won't be surprised if Apple uses your feedback for their upcoming Siri. Unfortunately based on this and other feedback here it looks like Martin had a very late launch. Hope this turns into a proper product because we do need a Siri competitor.
for what you're after, my app howie.ai would work well. we don't do nearly as much as Martin does, instead we're narrowly focused on getting the scheduling use case right. i'm a@howie.ai if you want to try it out!
An alternative for the background which might be slightly easier would be to use the same image as the vinyl, but apply a blur and scale up.
Other ideas for branding:
- Use the image to generate a colour palette to ensure text content has good contrast on light/dark images and is on-brand
- Allow users to add links to the track on Spotify, YouTube, Apple Music etc and generate a short random link (e.g vinlo.co/VUSNF) to use instead of the vinlo.co mark. You can use this to a) provide track links for end users, b) provide analytics to creators, and c) understand which creators are driving traffic
In your analogy it would be more like paying someone to mow your lawn because your neighbour got it done for $10, then being charged $100 because your house number is even.
It might be in the terms and conditions, but it’s bad faith to not give any warnings or controls before the services are rendered.
- Sideband: iOS/Android chat app (https://github.com/markqvist/Sideband)
- NomadNet: Desktop CLI chat app (https://github.com/markqvist/NomadNet)
- Rnode: Reference node hardware/firmware (https://unsigned.io/rnode/)