Hacker News new | comments | show | ask | jobs | submit login
Jupyter: “take the domain name down immediately” (sagemath.blogspot.com)
83 points by williamstein 211 days ago | hide | past | web | 38 comments | favorite

I don't know from the legalities of the situation, and it sounds like they don't matter to the author either, and they're just going to comply out of good-will.

But I will say that as a (very) casual user of Sage and someone with a passing knowledge of the Jupyter branding change: SageMathCloud having an alias of "JupyterCloud" is an instance where the name usage would be super ultra confusing to me. It would make me think SageMathCloud, a project I was already aware of (but not a user of), was the official cloud backend for Jupyter, and was just slow on the rename.

Many thanks for your feedback, which is exactly the sort of additional perspective I'm hoping for. I definitely don't want to confuse users, and I really appreciate your pointing this out. Yes, not a legal question: I will bend over backwords out of good will... but I'm not even sure what I'm asking for, which is why your feedback is so helpful.

Just to be as fair as I can: I also find the new "Jupyter" branding to itself be super ultra confusing. :)

This, anything with the word Jupyter in it causes confusion. Because homophones are cute, and supporting Ju(lia) Py(thon) and Ter makes sense, when the word IPython is too disclusive.

Technically there is also a (legal) issue with the IPython name and the proximity with `Python` trademark, where the IPython team have an exceptional authorisation, under continuous approval to use the name under some restrictive conditions, like for example non-commercial use.

> they're just going to …

He; William Stein is a he, not a they.


Thanks for pointing it out but I doubt this new information changes anything. It is grammatically correct to use singular they when I can't be bothered to find out the subject's gender.

It may be grammatically correct to the 51% of Wikipedians who can stomach working on that particular page, but it is highly annoying.

It is also impolite if the gender of the person in question is completely obvious.

Additionally, it is confusing for non-native speakers who always have to figure out whether something is plural or singular.

It might be highly annoying to a small percentage of HN readers, but so are many other things.

It's impolite to use a new, throwaway account to foist personal preferences rooted in 19th century prescriptive grammar on us all.

I have a hard time in German remembering if a noun is feminine, masculine, or neuter. They should get rid of the gender system to make things easier for me.


You might be wary about using that term. The use of the term "virtue signaling" is also a common method by which bigots signal their virtue to other bigots. As http://rationalwiki.org/wiki/Social_justice_warrior#.22Virtu... puts it, the term was:

> "seized upon by reactionaries almost immediately and broadened to just every instance of somebody publicly expressing a desire for a better world. Just like SJW and whiteknighting, the user assumes being a prejudiced, oppressive asshole is the default and knows saying as such won't get them any traction, so they throw terms around that assumes bad faith of "the other side" in order to poison the well. "

Since I doubt you have no desire for a better world, I'll assume you mean it in Bartholomew's original sense. Unfortunately, others have ruined it enough that it's no longer a meaningful term of discourse.


How do you know the downvotes weren't due to factors other than being correct?

For example, I can make a correct comment which has nothing to do with the thread. Irrelevant comments are infrequently downvoted. Eg, HN:ers downvote for reddit-style jokes. https://hn.algolia.com/?query=reddit%20joke%20downvote&sort=... . Even if there's nothing false in the joke.

People also downvote for tone: https://hn.algolia.com/?query=tone%20downvote&sort=byDate&pr... .

Since there are many factors which may cause someone to downvote, I don't think you can assume it was due to being correct.

Quoting from the HN Guidelines: "Please resist commenting about being downvoted. It never does any good, and it makes boring reading."

P.S. And how do you know 1928321931 prefers to be referred to as "he"?

No idea on legality, but personally, I find the whole naming around IPython notebook/ Jupyter notebook/Jupyterlab itself to be confusing enough, and if you were to have an official looking domain like JupyterCloud to redirect to a product called SageMathCloud, I'll just ended up thinking of SageMathCloud as the hosted commercial product of Jupyter, so yeah, it is confusing (and also I find it to be not a very nice/honest thing to do).

p.s. Unless you're BuzzFeed, clickbait title that looked like it is looking to stir up drama is just gonna make you look less trustworthy (which is a shame, since the article itself is about a pretty important topic and very much deserves a discussion).

Since this does seem to be a causing confusion, which is precisely the kind of confusion trademark law is supposed to prevent, it seems like changing the name is the right thing to do.

Would Jupyter also object to jupyter.sagemath.com? I mean, that seems to more clearly say that it's jupyter instances associated to sagemath.

jupyter.sagemath.com -- that's a really good idea. Thanks for pointing it out.

The title here was super clickbaity. The actual quote was "...we ask that you take the domain name down immediately, as it uses the Jupyter name"

William is looking to be hired by buzzfeed I guess. “take the domain name down immediately, and see what's happen after...”

You're right. I'm sorry the title sounded like linkbait, but it was the best I could think of to convey the topic, which was permission for use of a domain name with Jupyter in it.

Trademarks must be defended or they risk being lost. It seems to me (a legal layperson) that this is not only a reasonable action, but also, unfortunately, a necessary one.

Too many people use this as an excuse for enforcement behaviors that go well beyond what's necessary to maintain a trademark. It's possible, for example, to begin the conversation politely, offer to work together to find a way to properly license the mark for use, etc. without resorting to harsh-sounding demands. But nobody does this, and nobody who uses harsh demands (or threats, or other extreme tactics) ever faces any consequences because someone on the internet will defend them with a "well, trademark law required that, you know".

(disclaimer: I'm on the Jupyter steering council, but speaking unofficially...)

Very good point about wanting to work together. And we definitely do want to work together with people (and we have a good relationship with William in particular). Another relevant part of the message to William and to the community is:

"We are currently discussing in the steering council guidelines for proper use of the Jupyter name (started in part by your inquiry a few weeks ago). We hope to have such guidelines published soon. In the interim, we ask that you (and others) not use the name Jupyter in a domain name. We feel that this avoids confusion in the community and is being fair to all who have inquired about such matters."

So we (Jupyter steering council and community) are also trying to figure out a good policy - we (speaking generally) want to be conservative and avoid confusion until such policies are in place, though.

Here's a good example of C&D letter that isn't written with going on the "all out offensive" in mind: https://brokenpianoforpresident.files.wordpress.com/2012/07/...

Thanks. I really like that letter!

> without resorting to harsh-sounding demands. But nobody does this

The example in the article was polite, especially given the context that there was still some discussion to be had on Jupyter's side about the the issue.

"we ask that you take the domain name down immediately" isn't especially polite. It isn't incredibly rude, either, but I wouldn't hold it up as an example of a polite request.

Agreed. I wish we would have phrased it more congenially, especially given William's good relationship with the project already.

It didn't feel very polite; when I read the message I was outside picking blackberries with my 7-year old nephew, and "immediately" was impossible to comply with.

Yes, I'm sorry it was phrased as "immediately", which was unnecessarily harsh.

Which would make sense if the Jupyter project had a trademark, but it's not clear to me if they ever registered it. However, it would probably be considered a common law trademark.

It doesn't seem like they have registered it. Last year Fernando wrote that he hadn't (https://mail.scipy.org/pipermail/ipython-dev/2015-March/0162...) and searching TESS yields no results (http://tmsearch.uspto.gov/bin/gate.exe?f=searchss&state=4806...). That said, I am much more interested in what the Jupyter devs really want rather than any legal questions.

Could the trademark be licensed for $1 or something?

Could be licensed in exchange for a link on the front page to the Jupyter website; there is no requirement for copyright or trademark owners to offer licenses to all potential users for the same price.

Since Jupyter notebook is one type of things among others used in SageMathCloud, it's prudent to put it under the SageMath brand. You do want to build up the SageMath brand. Just put it under cloud.sagemath.com/jupyter.

Several people have now suggested exactly this to me too. I'll clarify my question to the Jupyter Steering Committee and also ask about use of Jupyter in a subdomain or page like cloud.sagemath.com/jupyter.

Frankly I think the name Jupytercloud is much, much better than sagemathcloud. The reason is that anything containing the word 'math' instantly kills 99.9% of your targetgroup. I have been a mathematician for half of my professional life and I share a a love for all things mathematics, but in the second half of my professional life I have learned (the hard way) that 99.9% of people instantly shut off as soon as math is brought up. Sadly this is even true for most engineers and scientists. Anything called mathcloud is instant negative marketing, especially as doing a google search quickly turns up, talk about L-functions, number theory and commutative algebra. All of these capabilities you are rightly proud of, but which for most anybody outside of academic math departments it is a clear sign that sage/sagemathcloud is "not for me". This includes mathematics teachers, who, however, seem to like geogebra, a system with much more limited capabilities, but sufficient for elementary algebra, plotting functions and elementary geometry. One of its strength is that it can used in a (chrome) browser.

On the other hand there is a much larger group (if only non-math students) for which a "google docs/github" for shared, zero-install, browser based python/R/Ruby programming, data-analysis, plotting, statistics, technical-scientific writing /Numpy numerical and yes, Sage symbolic computation environment is a great proposition. Of course this is exactly what SageMathcloud is offering. However, if I may exaggerate a bit, SageMathcloud is now positioned as a cloud system to support mathematicians that can also do things like plotting and statistics, because it is really handy for doing statistics of the zero's of the zeta function. Completely understandable from where you come from, but a marketing disaster. A name like Jupyther cloud that reflects such a "neutral" programming/technical/statistical/data position would be a great boon.

Why do you have to redirect jupytercloud to sagemathcloud? I feel that is what they are objecting to primarily. Create your web site with your own content for your domain and somewhere there, if needed or inclined to, provide a link to SageMathCloud. It will also help if you put appropriate disclaimers/clarifications so there is no confusion.

I did that redirect for testing purposes among the SageMathCloud developers; I did not advertise or link to it, and don't know how anybody found out about it [edit: somebody saw a github issue that mentioned the domain name]. The next step, which was being actively worked on the last two days by Tim Clemans, was modifying the site to change the content as you suggest. I'm too used to doing open source development very openly, and definitely this was a mistake in this case.

I think it was perfectly reasonable to grab the domain name as it could have been squatted by anybody else who would have been aware of this name. Just the fact that it resolves is/was problematic.

Off topic; hopefully this reaches the author: This is impossible to read on my iPhone 6. The text goes way off the side of the screen, but I can't scroll horizontally to read it. Also double-tapping the screen doesn't make it fit the text on, which usually works. Actually pinch-to-zoom doesn't work at all.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact