HNHacker News
TopNewBestAskShowJobs

jacobian

4,122 karma · joined December 8, 2008

http://jacobian.org/
submissionscomments
jacobian··on There's no magic: virtualenv edition
Related: Carl Meyer's "Reverse-engineering Ian Bicking's brain" talk from PyCon 2011: http://pyvideo.org/video/389/pycon-2011--reverse-engineering...
jacobian··on Ask HN: being offered stocks, how to estimate value?
I'm a consultant as well. I never take stock or equity as payment; cash only. Look at it this way: if the company's so cash-poor that they can't afford to pay for work, what are the odds their stock/equity will ever be worth anything?
jacobian··on _why's site is back up
If you hit `/SPOOL/{TEST,IDEAS,NOTES}` you get some binary data; `file` tells me it's "HP PCL printer data". Not sure what to do next with that...
jacobian··on Django 1.5 release candidate available
Well... 3 months is possible, but unlikely. 6-9 is more likely. 12 months is unlikely, but more likely than 3.
jacobian··on Django 1.5 release candidate available
It'll likely be a bit longer than that. See my response elsewhere in this thread (https://news.ycombinator.com/item?id=5009625).
jacobian··on Django 1.5 release candidate available
At some point, yes, we'll drop Python 2 support. When really depends on how rapidly we see adoption (from users and re-distributors of Django) and how heavy the dual-version support burden turns out to be. My guess is what we're looking at a longer support window than the 2-3 years you posit, but that's really just a half-assed guess. Anything between "next year" and "never" would be a reasonable guess too...

... and y'know, Python 3 should be on your radar, if not necessarily what you use right now. It's better than Python 2 in every way (except for adoption, and that's changing). 3.3 in particular has a ton of really great new stuff (futures! working namespace packages! yield from!) that you'll wonder how you lived without.

jacobian··on Django 1.5 release candidate available
So here's the deal:

Django 1.5 has "full" Python 3 support, at least from a technical level. It's done, works, is well-tested. I'm planning on putting a Django 1.5 / Python 3.3 site into production later this year. So in that sense it's "done" now.

However, we sorta have to hedge our bets in terms of what we recommend others do.

We have a pretty serious backwards-compatibility promise, and we're not 100% certainly we've nailed down all the Python 3 APIs perfectly. There's a small (but non-zero) chance something might need to change between 1.5 and 1.6; we want the ability to make that change backwards-incompatible (for Python 3 users) if possible.

Similarly, there's a small (but again non-zero) chance that there's a serious bug or three in the Python 3 support. We don't want people getting bitten.

Finally, much of the "good stuff" in Django-land is actually in third-party apps, many of which don't yet support Python 3. This puts a pretty big obstacle in the way of a Django/Py3 site -- most real-world sites use a ton of third-party stuff. So to use Django on Py3 right now you'll need to make a bunch of "should I DIY this component or port it to py3 and submit a patch?" decisions.

In the case of the site I mentioned up top these are tradeoffs I'm willing to make. I know Django well enough to not be worried about potential bugs or backwards-incompatibilities, and I have the luxury on this project of getting to work on py3 patches as part of the deal.

Most people won't be so lucky, though, so we expect the majority of Django users to wait until 1.6 to start seriously using Py3.

We don't have a firm timeline yet. I'm guessing something like 6-9 months for 1.6, but that depends on a lot of different things. It could be longer, as much as a year, or possibly even as short as 3-4 months. We'll see :)

jacobian··on The open-source "svblte like" blogging platform
Kick-ass; thanks!
jacobian··on Fab's Second Pivot
Actually, no, I'd really like for you to address this in public. You offered to answer any/all questions, right? The highest voted comment in this thread is https://news.ycombinator.com/item?id=4918790, which is the one that is accusing you of spam. That guy received the Aston Kutcher spam despite having opted out of everything (as did I); can you respond to him, please?

[BTW, you might want to be careful with the celebrity endorsement. I can't think of a celebrity I find more noxious than Kutcher, so attaching his name to Fab makes me much much less likely to buy anything from you again. Especially when you're forcing that association into my inbox. Blech.]

jacobian··on The open-source "svblte like" blogging platform
Hey simvla - looks good, but to actually be open source you need a license. Right now the source doesn't have a license, so it defaults to "All Rights Reserved". If you intended this to be open source as the title claims, can you add a license please? Thanks!
jacobian··on Fab's Second Pivot
Can you please respond to the accusations of spam, above? I, too, have suddenly started receiving email from Fab even though I've opted out repeatedly.

I really hope this is a mistake and not part of this third "pivot".

jacobian··on Python Software Foundation will require code of conduct at all sponsored events
Women in Denmark make about 10% less than men for the same work [1]. Granted, that's better than the US, where it's more like 75¢ on the dollar, but I'd hardly call that equality.

I don't know enough about Denmark to know whether gender discrimination persists in technical fields, but my guess is that women make up far less than half of the technical workforce (that's the case for most of Europe [2]).

[1] http://economix.blogs.nytimes.com/2010/03/09/the-gender-wage... [2] http://dl.acm.org/citation.cfm?doid=543812.543839

jacobian··on Python Software Foundation will require code of conduct at all sponsored events
Sheesh, there's just so much that's wrong, disingenuous, and outright hilariously silly about these couple of sentences that I hardly know where to begin. Let's see...

* You don't have a right to attend a private event. The organizers of any event - a conference, a concert, whatever - have the right to ask you to leave at any point for any reason. You are, in fact, more protected with a code of conduct in place because it articulates the process instead of leaving it up to the whims of the organizers.

* You don't have any right to due process or "legal protection"; the conference organizers aren't the police! They're not deciding what's "truth" or not; they're deciding whather to allow you to stay at an event they're responsible for.

* If you're talking about PyCon, it's hardly "a few local programmers". PyCon is a conference with a large staff, a budget of over a million dollars, and an established NPO (the PSF) behind it.

* If you can't trust the organizers of an event, then why the heck are you attending?

* People don't just randomly make up harassment. In reality, harassment is vastly underreported because of the way victims of harassment are routinely ignored, shamed, and blamed.

* Wanna know a really good way to not be accused of harassment? Don't harass anyone! If you find not being a harasser so hard, then yeah you probably shouldn't go to the event.

jacobian··on Python Software Foundation will require code of conduct at all sponsored events
> Living in a [...] gender-equal country [...]

Which country is that?

jacobian··on PyCon US 2013 Accepted Talks
If cost is the only thing keeping you from coming to PyCon, please considering applying for financial aid. Each year we give out tens of thousands of dollars in free tickets and grants to help people get to PyCon. You could easily be one of them - please apply! See https://us.pycon.org/2013/assistance/ for more details.
jacobian··on PyCon US 2013 Accepted Talks
Yes, all the talks will be recorded and release after the conference.
jacobian··on PyCon US 2013 Accepted Talks
Great idea - I'll see what I can do!
jacobian··on PyCon US 2013 Accepted Talks
Most talks will be 30 minutes, some will be 45. We don't know which are which yet; that'll come when we publish the actual schedule in the next couple weeks.
jacobian··on PyCon US 2013 Accepted Talks
If y'all have questions, I'm the program chair so fire away. I'm happy to answer 'em here.
jacobian··on Django is falling behind
Funny, your TODO list looks a lot like mine.
jacobian··on Python Web Frameworks' Complexity
So Django doesn't have its own package management system, either.

I'm curious: where did you read/see that?

jacobian··on Python Web Frameworks' Complexity
Sorry, I didn't mean to suggest he was making up a metric.

Yes, McCabe is a real thing. I haven't really found it that useful myself -- I'm not sure that you can adequately measure something fuzzy like "complexity" through static control flow analysis -- but sure, I'm willing to accept that it could be useful in certain circumstances.

What I am saying is that this guy is cherry-picking a specific metric (out of many that he could choose) that makes his tool looks good. I could just as easily come up with a metric that sorts Django to the top of an arbitrary list. That wouldn't be useful, either.

jacobian··on Python Web Frameworks' Complexity
Note that this article is by the author of wheezy, a framework that just happens to do quite well on this particular metric.

Yes, I'm a maintainer of Django, so I'm biased too. But look: there's a good argument to be made that parts of (even of all of) Django is too complex. This, however, is not that argument; this is just a made-up metric.

ETA: I misspoke above: McCabe isn't made up. I meant to write that this was a cherry-picked metric. Leaving the above as is because people have already replied to it.

jacobian··on Django security releases issued
> However, my idea is just what's already outlined above: avoid using 'Host', except (perhaps) as a key.

Yes, this is a good idea in theory (it's an offshoot of the idea of avoiding using any data that's provided by the browser.) But in practice, it breaks down any time you need to generate a full URL into the site, i.e.:

* Callback URLs for schemes like OAuth, webhooks, IPN, etc. * Transaction emails with links back to the site ("this week on example.com", "joe just mentioned you..." password resets, etc.) * Links in Atom, RSS feeds, etc.

In each case, Django could require that hosts be hardcoded (perhaps in settings) instead of consulting requests, but that would make each of those things harder to use.

... I'm not really trying to have an argument here; I see your point completely and I think I probably agree, personally. But I am trying to point out that security and usability are often at odds, and that there isn't as much a "right answer" as there is a calibration between those two needs. Drawing the line in the right place is hard.

jacobian··on Django security releases issued
It's a tough line to draw. Requiring users to whitelist every domain that might be used to access the app is a PITA, and makes the lives of new developers more annoying. (It also makes multi-tennancy by domain name, e.g. customer.example.com, a bit more difficult.)

On the other hand, you're completely right that the "you gotta get your upstream server configured correctly" advice is handwavy at best.

Seems like this is always the case, this tension between security and usability...

If you've got bright ideas, I'd love to hear 'em. Maybe join us on django-dev (http://groups.google.com/group/django-developers) if you do?

jacobian··on Django security releases issued
> The fix as described seems to just check the Host header for suspiciously-formatted information. Is that also enough to prevent other wrong (but well-formatted) Host values from being used?

No, it's not. As the release notes say, "Some attacks against this are beyond Django's ability to control, and require the web server to be properly configured"; see https://docs.djangoproject.com/en/1.4/topics/security/#host-... for details.

jacobian··on Django security releases issued
Yup, that's basically it. It essentially makes phishing Django sites easier. The vulnerability could allow an attacker to send a legit email -- sent by the real site itself -- with a link that sends a user to a malicious site instead.

[So please upgrade!]

jacobian··on Grove isn't dead, it's only resting
Thanks for the feedback; let me try to give you some more info:

1. Yes, we're (painfully) aware that restoring users' trust is going to be a big deal. There's obviously no way I can prove the future, so all I can say is that we're in this for the long haul. Grove fits in perfectly with our business, is a product that we fundamentally believe in, and so it's not going anywhere. We're not going to sell it, or shut it down, or "pivot" into a different business. I hope I have a good enough reputation that you can take me at my word, but whatever: our actions over the coming months and years will prove this.

2. We're not re-branding; that feels dishonest (plus, I love the brand.) I'm confidant that we can restore whatever trust we've lost.

3. Oh hells yes, I want an awesome mobile site/app, too. Realistically it's a bit down the list after some of the backend stuff we need to hammer our first. But I want one, bad, so it'll happen as soon as we can!

Thanks again for the feedback, it's really helpful. If you've got anything else you'd like to tell me/us -- or if there's anything more I can do to get you back as a user -- please drop me a line (jacob at jacobian.org).

jacobian··on Grove is here to stay
Indeed, as Grove users the stability issues annoyed the heck out of us, too; fixing them is our first priority. As in, I'm already working on it!

I know migrating (again!) is a pain, but if you ever want to give Grove another try, please get in touch (jacob at jacobian.org) and let me know what we'd need to do to get you back.

jacobian··on Grove isn't dead, it's only resting
Yup, we certainly have some work to do to win users back. If you've got any suggestions, or if there's anything we can do to get you back, please let me know! jacob at jacobian.org.
← PreviousPage 4 of 10Next →