Servo’s New Home (2020) | Hacker News Reader

Servo’s New Home (2020)

servo.org

59 pointsAndrewDucker33 comments

Hmm, seems like the linux foundation is a nice retirement home for ex corporate projects that are popular enough that people feel sad seeing them die, but not actually popular enough to live. (See RethinkDB)

And aren’t written in Java (That’s Apache…)

How far is servo from a useable browser? What is it missing?

Well, there is an open issue for “opening servo.org crashes servo”

Now that's pretty funny.

That is its default homepage now I just downloaded it to try. Works fine. Placement is off for some items, but i haven't found a page that simply doesn't work. I poked around wikipedia mostly though which is fairly benign. Both google and duck duck go jad search suggestions working. It isn't convenient, but it isn't hopeless either.

Last I heard the goal of servo isn't to have an actual browser, but basically build an engine that can be used somewhere else and that all the "good stuff" from Servo goes into Firefox, basically.

That was the goal under Mozilla. Now that they aren't paying for the work anymore, I would expect other employers (and volunteers) to want more out of it than a tech incubator for another project.

It seems like that may already be taking place as part of this launch. I downloaded it to try after asking this question and it mostly works. Just some aesthetic stuff. If I can read news and wikipedia I'm pretty happy with a browser. Maybe make everything beyond documents a PWA or app?

Yeah, I'll admit my info/perception is out of date but I guess because after they split from Mozilla I mostly stopped following any updates. Maybe I'll play around with it this weekend just to see what it's like.

Interesting. Igalia is going to have 4 people working on Servo starting this month: https://people.igalia.com/mrego/servo/igalia-servo-tsc-2022/

> Servo’s mission is to provide an independent, modular, embeddable web engine, which allows developers to deliver content and applications using web standards.

What is the business use case for this?

As in, if each of those people costs $100k/yr, that's $400k/yr+

How will this product make (whoever is paying for it) money (profit)?

If that answer is "it won't, not everything needs to make money", that's totally fine. Could you then explain why this company is willing to shell out at least $400k/yr to have 4 people work on it? What do they get out of it (if not money)?

Might want to read the company's profile for more info: https://www.igalia.com/about/history

They are an open source software consultancy.

"Since 2010, Igalia has become the world’s leading web engines consultancy, and has been a top contributor to the WebKit and Chromium projects. Thanks to our excellent work, Igalia has since been accepted as a member of associations including the W3C, GENIVI, Automotive Grade Linux, the Khronos Group, and ECMA International."

But off the top of my head with their focus on embeddable IOT devices, that alone is a good business case. Having CSS/HTML available to render to OLEDs, eink displays, smart watches, etc, would be super nice.

What does this buy for your IoT use case that Chromium doesn’t?

Chromium doesn't run on most embeddables and is very memory and power hungry.

How big is the market for people running watered down web browsers on embeddables? Do you have any example usecases off the top of your head?

Sorry, I'm not going to write a whitepaper for you, you'll have to do that research yourself if that's a field you're interested in getting into. I already mentioned smart watches, but could be e-readers, e-ink tablets like Remarkable, diagnostic screens, etc.

Less bloat? How easy is it to strip Chromium down if you only want HTML+CSS for instance? (I would guess: not trivial)

They have mid-term goals at https://people.igalia.com/mrego/servo/igalia-servo-tsc-2022/... : "Web engine for some narrow use cases in small embedded devices (IoT)"

Given Igalia work on Webkit embedding, looks like they want to see if Servo could be a viable alternative in some contexts.

So sad that something useful only gets done when there is a business case for it.

From 2020, apparently.

Yeah, I clicked while murnuring "Again?" but this is old news.

What was the status of Servo WRT suitability to replace Gecko? I vaguely remember that some of the parallel layout techniques didn't work out. Is Servo seen as a dead end or just too much work to get production ready?

cmiiw the goal wasn't about replacing Gecko, but to incorporate ideas/components from Servo into Gecko itself

That's becasue the other goal would be way longer term. But it's not a bad goal to use a complete replacement if they could get resources for it.

How many people are working on Servo lately?

It's busier than I would've thought.

https://github.com/servo/servo/pulse/monthly

Almost all of those PRs are made by bots for updating dependencies

https://github.com/servo/servo/graphs/contributors?from=2021...

Most of the commits are updates to dependencies and the like.

So basically, no one is, outside of preventing too much "bit rot" and some very occasional bugfixes.

Discussed at the time:

Servo’s new home - https://news.ycombinator.com/item?id=25125325 - Nov 2020 (309 comments)

I wonder how hard it'll be to implement Android's WebView on top of Servo?

There's nothing new about this. It's from 2020.

Servo’s New Grave

Reply on news.ycombinator.com