[HN Gopher] Pydantic Logfire
       ___________________________________________________________________
        
       Pydantic Logfire
        
       Author : ellieh
       Score  : 89 points
       Date   : 2024-04-30 15:56 UTC (7 hours ago)
        
 (HTM) web link (pydantic.dev)
 (TXT) w3m dump (pydantic.dev)
        
       | serjester wrote:
       | I love pydantic but I really have to wonder why they chose this
       | route. There's already a ton of companies that do this really
       | well and I'm trying to figure out how their solution is any
       | different.
       | 
       | The llm integration seems promising but if you care about LLM
       | observability you probably also care about evals, guardrails and
       | a million other things that are very specific to LLM's. Is it
       | possible to build all this under one platform?
       | 
       | I do hope I'm wrong for the sake of pydantic-core.
        
         | thenipper wrote:
         | Agreed. I use pydantic daily and I'm in the position to try
         | this out... but why? It really seems to be a weird divergence
         | from their core product.
        
           | Yenrabbit wrote:
           | They also have a UI framework (fastui). They have funding and
           | devs looking to be useful, I think it's fine that they get to
           | try out new things that work well with their core product.
        
         | wferrell wrote:
         | Who do you like who does this well?
         | 
         | I think Pydantic is great software and so I am inclined to see
         | if this too will be great software.
        
           | clintonb wrote:
           | My last company used Splunk and SignalFx. They were fine, but
           | I wasn't necessarily in the weeds with configuration and
           | usage. The Observability Team made it easy to use.
           | 
           | My current company, a much small startup, primarily uses
           | Datadog and we are starting to better integrage Honeycomb. We
           | mostly abandoned Google Cloud Monitoring because the UI/UX
           | are not that great. Honeycomb is a paradigm, so took some
           | time for us/me to understand. It's growing on us.
           | 
           | Despite the ability to completely blow up our bill (which is
           | now better controlled), Datadog is a good product that lets
           | us quickly answer questions when things go wrong. It's not
           | perfect, but it's the best we have right now. The UI is
           | intuitive, and facilitates more discovery (esp. for metrics
           | and their attributes).
        
             | cipherself wrote:
             | I am kinda confused by your comment, OP is about tracing as
             | far as I understood, but you're referencing Google Cloud
             | Monitoring (whereas the comparable thing would be Google
             | Cloud Trace) and then again                 (esp. for
             | metrics and their attributes).
             | 
             | but OP isn't about metrics at all rather traces.
        
               | gazpacho wrote:
               | Hi, I'm one of the developers of Logfire. We do support
               | metrics! Our vision is a single pane of glass
               | observability where you can easily go from metrics <->
               | traces both via correlation and by creating
               | metrics/dashboards from traces.
               | 
               | We also support logging as an integrated concept into
               | tracing (you can emit events without a duration that are
               | just like a log but carry context of where they were
               | emitted from relative to a trace).
        
           | emmanueloga_ wrote:
           | Looks like "Pydantic Logfire" is another entry on the
           | category of "APM"s? [1]
           | 
           | Gotta echo the sentiment that Logfire doesn't seem to be too
           | closely related to Pydantic... Also, afaict it looks like the
           | frontend is not open source, unless I'm missing something
           | [2]. So, not a tool that one could self-host?
           | 
           | --
           | 
           | 1: https://github.com/topics/apm
           | 
           | 2: https://github.com/pydantic/logfire/
        
         | scolvin wrote:
         | Pydantic creator here.
         | 
         | I understand why this might be your reaction, but let me just
         | share my thoughts:
         | 
         | Can we build all the LLM things people want? Yes I think so,
         | early feedback is that Logfire is already much more
         | comprehensive than LLM specific solutions.
         | 
         | How is our solution any different? AFAIK:
         | 
         | * no one else offers opinionated wrapper for OTel to make it
         | nicer to use, but with all the advantages
         | 
         | * no one else does auto-tracing (basically profiling) using
         | import hooks the way we do
         | 
         | * no one else has dedicated instrumentation for OpenAI or
         | Pydantic
         | 
         | * no one else provides metadata about python objects the way we
         | do so we can render a semi-faithful repr of complex objects
         | (Pydantic models, dataclasses, dataframes etc.) in the web UI
         | 
         | * no one else (except Sentry, who are doing something
         | different) makes it really easy to start using it
         | 
         | * no one else lets you query observability data with SQL
         | 
         | In my mind, the obvious alternative would be to do the thing
         | most OSS companies do, and build "Pydantic Cloud", then start
         | adding features to that instead of the open source package. I
         | didn't want to do that, and I don't think our users would like
         | that either.
         | 
         | In the end I decided to build Logfire for two reasons:
         | 
         | 1. I wanted it, and have wanted it for years
         | 
         | 2. I think building a commercial product like this, then using
         | OSS to spread the word and drive adoption is a really exciting
         | way to incentivize us to make the open source as good as
         | possible -- good as-in permissive and good as-in well
         | maintained. And it means that our commercial success is not in
         | tension with the adoption of our open source, which has been a
         | recurring issue for companies trying to capitalize on their
         | open source brand.
        
           | hakanderyal wrote:
           | Useful product, great marketing site. This is obviously a
           | labor of love. Keep it up!
           | 
           | Sadly I've moved on from Python world.
           | 
           | Funny anecdote, using Pydantic everywhere to improve
           | maintainability made me realize I'm fighting an uphill battle
           | with Python and I should move to a statically typed language,
           | so I switched to C#.
           | 
           | Thanks for your work.
        
           | serjester wrote:
           | Thanks for the response, massive kudos - I look forward to
           | trying it out!
        
           | btown wrote:
           | This is very cool. Could someone use Logfire alongside
           | Honeycomb or another OTel platform? Honeycomb's strengths in
           | drilling into an interactive chart vs. Logfire's strengths in
           | drilling into a stream of text logs (and SQL support!) seem
           | very complementary.
           | 
           | Certainly seems like the Logfire initializer could coexist
           | with code like in https://docs.honeycomb.io/send-
           | data/logs/opentelemetry/sdk/p... - is this a reasonable
           | assumption? We're looking into modernizing our Honeycomb
           | integration from their older direct integration to their OTel
           | SDK, and would be very curious if Logfire can exist alongside
           | this. Do you do any patching of OTel internals that would
           | make this impossible?
        
             | scolvin wrote:
             | You can send data from any OTLP endpoint, so yes.
             | 
             | I don't think we currently expose an easy way for you to
             | send data to two (e.g. honeycomb and logfire), it it should
             | be entirely doable, and if people want that, we can make it
             | easy.
        
       | zbentley wrote:
       | This smells a bit like someone is using the brand of a famous
       | open-source tool they built to promote their startup. If so,
       | that's lame; if the new business is unrelated to your open source
       | project, branding it as _Pydantic_ Logfire seems a bit
       | disingenuous.
       | 
       | I think there's a less-misleading way to use open-source
       | reputation for business credibility, though: "so-and-so business,
       | by the creators of so-and-so project". Knowing that respected and
       | skilled folks are working on a business is great! It's a fine
       | distinction, but I think it matters.
        
         | scolvin wrote:
         | If anyone is mislead about whether Pydantic is Logfire or visa-
         | versa, I'll eat my hat.
         | 
         | Other than that, see my answer to the other comment.
        
           | nsteel wrote:
           | For the record, I don't agree with the grandparent comment at
           | all. None of this is "lame". It's a path to sustainable OSS
           | and it's more obvious than ever that we need these pathways.
           | 
           | But to be fair, the branding here is a little weird. If you
           | were living under a rock and hadn't heard of pydantic then
           | the website reads like pydantic is the company and logfire is
           | a product of said company. That's fine. But then you've also
           | got a product called pydantic. Or is it now called pydantic
           | pydantic? I realise it's kind of an extension to pydantic but
           | the AI focus doesn't make it feel that way and so I think
           | that's where the GP is coming from. Sorry to nitpick, I hope
           | it does well.
        
             | esafak wrote:
             | This is just how you brand when you have one product. If he
             | made up a new brand to subsume both products nobody would
             | recognize it, so he wisely decided to recycle the existing
             | brand; familiar, and liked.
             | 
             | Google was originally just the search engine, without any
             | qualifier. Now it's Google Search. In the same way,
             | pydantic can become "Pydantic Validate" or something.
        
             | scolvin wrote:
             | I think that's fair. It's a hard problem.
             | 
             | We could have called the company a completely new name and
             | everyone would have been confused, or just called us
             | "pydantic".
             | 
             | I think Pydantic being the company name, and a standalone
             | entity, and there being other products is fairly common. I
             | think if Logfire is successful, it will end up just being
             | known as "Logfire".
        
         | esafak wrote:
         | It is related, though: the creator is the same, so you can
         | expect the same DX.
        
         | robrenaud wrote:
         | Finding ways to reward people for building great open source
         | software is something that we need more of. It's not lame at
         | all.
        
         | barapa wrote:
         | From the website:
         | 
         |  _From the team behind Pydantic, Logfire is a new type of
         | observability platform built on the same belief as our open
         | source library -- that the most powerful tools can be easy to
         | use._
        
       | MapleWalnut wrote:
       | It'll be hard to position this against Sentry. Sentry's a joy to
       | use and their performance product is so helpful in debugging
       | performance issues
        
         | Lucasoato wrote:
         | One of the Sentry inconvenience is self-hosting: it relies on
         | so many services it can be very complicated to maintain.
        
           | esafak wrote:
           | Does the self-hosted version have all the features?
        
             | easton wrote:
             | yes, every feature (but billing, iirc) is in self hosted.
             | which means you need all of their backend dependencies
             | (kafka, zookeeper, other things that are probably easier to
             | manage than those two)
        
           | mdaniel wrote:
           | I draw ones attention to the _actual_ Open Source glitchtip
           | which has a much more sane deployment, akin to the good old
           | days of Sentry before they got Big Data-itis:
           | https://gitlab.com/glitchtip/glitchtip-
           | backend/-/blob/v4.0.8... (or its helm version, similarly not
           | JFC https://gitlab.com/glitchtip/glitchtip-helm-
           | chart/-/tree/61c... )
        
       | drchaim wrote:
       | The ideal business model (from a user's point of view) for
       | successful libraries like Pydantic, Ruff, etc., is like a
       | foundation or sponsoring. But (almost) nobody pays for them (me
       | included), so their creators have raised VC money, citing: 'We
       | have xxM users and yyyyyM downloads per month.' Not to blame
       | library authors here, they are trying their thing, which is
       | indeed admirable. It's just that open source is difficult to
       | sustain. Wishing them all the best
        
       | jmduke wrote:
       | OP: as someone who runs a Python business and is a happy user of
       | Pydantic (via Django-ninja) -- something I would want this splash
       | page to convey, at least implicitly, is "why use this over
       | Sentry?" As far as I can tell, the answer is: an OpenAI binding
       | and stronger Otel support.
       | 
       | (That being said, I am a sucker for all new tooling in this genre
       | and am excited to play around with it!)
        
         | jmduke wrote:
         | Responding to my own post: this comment
         | (https://news.ycombinator.com/item?id=40214695) I think is a
         | very compelling answer. I love the design language of the
         | splash page, but I think the copy could be improved by a little
         | less "here is why performance insights are useful" (which I
         | think most people are kind of already on board with) and more
         | "here is why our performance insights are _particularly_ more
         | useful than the competition".
        
       | megaman821 wrote:
       | Does this do logging or just traces? More to the point, is it
       | economically to store tens of millions of log lines a month?
        
         | cipherself wrote:
         | From the screenshots, it seems very much to be about tracing.
        
         | dmontagu wrote:
         | I'm one of the Pydantic/Logfire devs -- we are actually aiming
         | to unify the ideas of logging and tracing a bit more than seems
         | common in most such tools -- our logfire SDK actually emits
         | spans for all of the traditional logging statements rather than
         | relying on OTel logging infrastructure (which seems to be kind
         | of a legacy-oriented afterthought relative to the tracing
         | APIs/infrastructure). So, yes, I would say we do support
         | logging (certainly on a conceptual level), though the point is
         | to try to move toward a world where as much as possible is done
         | with spans, if only as an implementation detail.
         | 
         | We also have explicit integrations with popular logging
         | packages (see the bottom of https://docs.pydantic.dev/logfire/i
         | ntegrations/#opentelemetr...), so if you use the standard
         | library `logging` module, `structlog`, or `loguru`, you should
         | be able to set up your logs to ship to Logfire with minimal
         | modifications to your existing code.
         | 
         | While we have not announced our pricing yet, I am confident
         | that tens of millions of log lines a month will absolutely be
         | economical to store.
         | 
         | (Also, I'll just add that we also support metrics, though our
         | metrics support is not as well-integrated yet as tracing. But
         | soon!)
        
           | laborcontract wrote:
           | I was just looking for a logging service. Really enjoy
           | Pydantic so this looks like it may be a delight.
           | 
           | Try not to surprise us on pricing, though.
        
       | barapa wrote:
       | I'm already using it and I think it's great. Well done Samuel and
       | team!
        
       | logfire_q wrote:
       | can i self host logfire? based on the "logfire auth" in the docs
       | i'm thinking no?
        
       | mariocesar wrote:
       | It is a great tool! Congratulations on launching it!
       | 
       | Could you clarify if it's possible to self-host the service?
        
         | scolvin wrote:
         | We'll offer self hosting for enterprise.
         | 
         | I want the entry level for "enterprise" to be much lower than
         | some other companies in this space.
        
       | mulmboy wrote:
       | Is self hosting possible? Can't see any mention in the docs
        
         | ryanisnan wrote:
         | It looks like it's on the roadmap, but my interpretation of the
         | project would be that this is would be behind a licensed
         | paywall. https://docs.pydantic.dev/logfire/roadmap/#on-premise-
         | deploy...
        
       ___________________________________________________________________
       (page generated 2024-04-30 23:00 UTC)