First top level ASF graph project

[ThomasThelen]

The main website says that Huge Graph is

The First Apache Foundation Top-Level Graph Project

Huge Graph was accepted 2026 and there's a longer history of other Apache top level graph projects: Apache AGE (2022), TinkerPop (2016), Giraph (top level since 2012 ), or Jena (top level in 2012).

What is meant by the website?

[byteayan]

You're right, the banner doesn't hold up as written. Jena and Giraph became top-level projects in 2012, TinkerPop in 2016 and AGE in 2022, all before HugeGraph graduated.

The line comes from data/home/en.yaml in apache/hugegraph-doc (tlp.title), and the Chinese page has the same claim (首个 Apache 基金会的顶级图项目). I'm not a maintainer, so I can't say what was meant. It might have meant something narrower, such as the first graph database with its own distributed storage to graduate, but the sentence doesn't say that.

I can open a PR on hugegraph-doc to change both pages to something like "An Apache Software Foundation Top-Level Graph Project", unless a maintainer would rather keep the claim and make it more specific.

[imbajin]

Thanks @ThomasThelen and @byteayan. I missed the original question and only noticed it with the recent reply. The wording should be clearer.

What we intended to convey was “the first full-stack graph system at the Apache Software Foundation.” The emphasis is on the integrated system: graph storage and querying, graph computing, and graph AI, together with the supporting toolchain. This matches the homepage introduction: “HugeGraph is a full-stack graph system covering graph database, graph computing, and graph AI.”

Other Apache graph projects have a longer history and different areas of focus. Our intended distinction was the combination of these capabilities within one project, rather than a claim to be Apache’s first graph project.

Would “The first xxx graph system" at ASF communicate this clearly? I’m not sure & I’d welcome suggestions, including wording without it. The main thing we want to capture is HugeGraph’s full-stack scope. Happy to discuss and adjust the description.

[byteayan]

Thanks @imbajin, that makes sense. I'd drop "first" altogether, since any "first X" claim invites the same question. Something like:

• EN: A Full-Stack Graph System at the Apache Software Foundation
• CN: Apache 软件基金会的全栈图系统

It matches the homepage intro and still puts the full-stack scope up front. If that works for you, I can open a PR on hugegraph-doc updating data/home/en.yaml and data/home/cn.yaml.

[ThomasThelen]

Thank you both for clarifying!

I like @byteayan's idea above although it would be awesome to see something specific and shiny unique to HG for prospective users. I removed the AFS reference because I think it's usually an afterthought (this is my own bias). The idea here is to have a hook letting everyone know that HG can solve their graph & compute needs. When the light bulb turns on in their head, then they look in to governance and see that it's within ASF.

A Full-Stack Graph Storage and Compute Engine

I'm not a huge fan of the full-stack bit. I see what's being conveyed now but someone looking for a graph database/engine probably won't know what's being implied

A Massive Graph Storage and Compute Engine

If you want the AI angle,

A Massive Graph Storage and Compute Engine for AI & Analytics

I think that playing on semantics or finding a qualifier to make first work can be a slippery slope - especially if we're two qualifiers deep (First X in Y) where the second is already niche (ASF). To me the optics look a little reachy; picking a tag line is incredibly difficult and impactful for users finding a project.

Then again, this is towards the bottom of the page and doesn't add much that hasn't been said in the top banner anyway. I think something around the number of projects or orgs that use HG would be a more enticing metric for users.

[byteayan]

Fair point about leading with what HugeGraph does rather than where it lives. Of your options, "A Massive Graph Storage and Compute Engine for AI & Analytics" reads best to me. It says what "full-stack" was trying to say without the reader having to decode it.

@imbajin, if that wording (or another one you prefer) works, I'll open the hugegraph-doc PR for en.yaml and cn.yaml. For the Chinese line I'd rather use your translation than guess at one.