Disqus - Latest Comments for jsmarr

Re: Full video of my Google I/O talk now available

jsmarr — Thu, 27 Aug 2009 02:28:54 -0000

Sure, it's Sxipper (sxipper.com), a great Firefox plug-in made by Dick Hardt.

Re: Programming wisdom (Scripting News)

dai_vernon — Sun, 16 Aug 2009 21:07:44 -0000

This is exactly what I believe.

Re: Programming wisdom (Scripting News)

jsmarr — Sun, 16 Aug 2009 20:38:48 -0000

Well said, Dave, the part about caring for the web while you're able to make an impact really stuck with me. At least I derive some hope from the increased focus on contributing open source software and supporting open standards where possible. In both cases, most companies have found a "selfish motive" to support this work that is also very good for the web at large (e.g. others will help support and improve it, it'll help us hire people, it will lower barriers to building integrations with others, etc.) but that came after the advocacy of people who could tell it was just very good for innovation in general to have companies do this. Hopefully the same can happen more broadly for things like URL shorteners!

Re: Tasted even better than it looks (Scripting News)

jsmarr — Thu, 06 Aug 2009 00:33:35 -0000

Sushi Ran is possibly my favorite sushi spot in the entire bay area. Every time I find myself north of SF, I try to sneak in a Sushi Ran visit on the way back. :)

Re: Status Update: Plaxo and Facebook are now in an “Open Relationship”

Silona — Wed, 29 Apr 2009 03:11:20 -0000

Are those gonna mesh w the facebook ones - cause that would make my life a bunch easier!

and um again proving the point of syncing information sources ;-)

so happy about this! Congrats !

Re: Status Update: Plaxo and Facebook are now in an “Open Relationship”

mikealexander — Tue, 28 Apr 2009 15:57:51 -0000

I have to say that surprises me. I would have thought Facebook would provide something that "pushes" status updates to authorised third party services as and when it changes.

Up to now, I've used TwitterFeed to update Twitter using an RSS feed from Facebook. The latency here is 30 minutes maximum. Do you think Plaxo's status sync with Facebook (and Twitter from there) will match, or beat that latency?

Re: Status Update: Plaxo and Facebook are now in an “Open Relationship”

jsmarr — Tue, 28 Apr 2009 13:50:21 -0000

We do offer user-created groups in Plaxo, so you can make a group and invite the subset of people you want to share with, and then when you share things in Plaxo, you can choose to just share with members of that group. For instance, we have a "Plaxo Employees" group that we use all the time to post articles/etc. for internal consumption.

Re: Status Update: Plaxo and Facebook are now in an “Open Relationship”

jsmarr — Tue, 28 Apr 2009 13:47:15 -0000

Yeah, we have to poll FB to look for updates on their side, so it sometimes takes a while (and we have a LOT of people hooking up their accounts right now, so not surprised our crawlers are a bit behind). Hopefully we can reduce that delay over time though. And updates from Plaxo to FB should happen right away because we can push them directly.

Re: Twitter and OAuth, interesting brew (Scripting News)

dave — Mon, 20 Apr 2009 19:47:26 -0000

That's good to hear. It would help simplify things for developers and give
more choice to users.

Re: Twitter and OAuth, interesting brew (Scripting News)

jsmarr — Mon, 20 Apr 2009 18:29:41 -0000

What's funny is that Facebook's proprietary delegated-auth system (fbauth) is almost identical to OAuth, so it *should* be super-trivial for them to support OAuth alongside fbauth as a way to access all of their APIs. And, IMO, it would be a great bang-for-the-buck developer and PR win for them, with no downside risk (since it's just replacing an existing component with a more standard and functionally equivalent version). Seems like some fb engineer that gets inspired should just hack this up over the weekend or something...any takers? :)

Re: Portable Contacts and vCardDAV (IETF 74)

jsmarr — Thu, 26 Mar 2009 14:40:34 -0000

Yeah, great seeing you there, and congrats on the amazing progress you're making with drizzle!

Re: Portable Contacts and vCardDAV (IETF 74)

jsmarr — Thu, 26 Mar 2009 12:02:22 -0000

Mike-they're complementary, and they play very well together. PoCo is an API with a contact info schema that's heavily based on vCard. And hCard is a way of embedding vCard data in web pages that don't otherwise have APIs. So for instance, if you have hCard for your contact info on your own blog, we can run that through technorati's hCard->vCard converter and then run that through our vCard->PoCo converter and voila--your home page is now effectively a Portable Contacts API for your own contact info (see for instance http://bit.ly/longpipe which does this with twitter).

Re: How FriendFeed uses MySQL to store schema-less data - Bret Taylor's blog

jsmarr — Fri, 27 Feb 2009 15:22:20 -0000

Bret-great article. We use a lot of the same tricks at Plaxo these days--most of our tables these days have a binary blob where we store compressed JSON for schema-less data, for exactly the same reasons you mention (some of our tables literally have billions of rows, and we faced all the same paralysis or ops-slave-swap-craziness to touch indexes, so we've tried as much as possible to move away from that). One insight that led us here, which you didn't mention explicitly in your post: SQL is designed for flexibility and arbitrary selects and joins across any column. But in practice, most apps like ours won't ever need to query or join on most columns. So there aren't really that many advantages to using *columns* for those attributes in the first place, at least not in DB terms. While we haven't gone as far as 100% separate tables for indexes, we do try to only use separate columns for fields we expect to index and query on, and shove as much as possible into these JSON blobs. The only downside is sometimes for internal stats purposes or fixing old/bad data, it would be nice if SQL could natively work with JSON-data in columns (seems like a reasonable thing to add in this day and age), but a little clever regexing in our queries handles most of our needs there, and it's wonderful to just be able to easily read and write arbitrary attributes into the JSON from PHP code, since the native data structures match up so perfectly.

Thanks again for sharing--we should all be doing more of this! js

Re: Missing the Oscars Finale: A Case Study in Technology Failure (and Opportunity)

jsmarr — Mon, 23 Feb 2009 17:43:11 -0000

Looks like facebook is also trying to get in on the act, albeit with a somewhat limited integration right now (live status updates embedded in some popular entertainment sites): http://blog.facebook.com/blog.php?post=55313427130 -- still I bet even that goes a long way for groups of people that keep that page up while watching the oscars and have several friends doing the same!

Re: Missing the Oscars Finale: A Case Study in Technology Failure (and Opportunity)

jsmarr — Mon, 23 Feb 2009 17:39:52 -0000

I just came across another relevant post on Yahoo's main blog a few days ago: http://ycorpblog.com/2009/02/19/the-envelope-go... -- see also http://ysearchblog.com/2009/02/20/get-ready-for...

Re: Missing the Oscars Finale: A Case Study in Technology Failure (and Opportunity)

jsmarr — Mon, 23 Feb 2009 14:52:39 -0000

Heh, just saw this link from Simon Willison. Sounds like progress in the right direction! :) http://simonwillison.net/2009/Feb/23/oscars/

Re: Implementing OAuth is still too hard… but it doesn’t have to be

dave — Wed, 18 Feb 2009 07:30:12 -0000

Why don't we have a dinner in March to discuss?

Re: Implementing OAuth is still too hard… but it doesn’t have to be

jsmarr — Tue, 17 Feb 2009 23:38:12 -0000

Ryan-sounds great, and I'll definitely take you up on that offer! ;) It's clear to me that a little extra time and effort by people like you and me whose jobs and companies depend on OAuth being easy to make it so will be a wise investment indeed.

Re: Implementing OAuth is still too hard… but it doesn’t have to be

daveman692 — Tue, 17 Feb 2009 22:23:01 -0000

A bit like your OpenID recipe that you wrote a few years ago. We're planning to write this sort of guide for our book, but haven't started on the OAuth content yet.

Re: Implementing OAuth is still too hard… but it doesn’t have to be

jsmarr — Tue, 17 Feb 2009 22:16:24 -0000

Actually I think the first step should be a good recipe guide and a companion "transparent provider" that the recipe uses. It would take the user through getting/writing a library, hitting a basic "hello world" up to a more complex API call, and it would provide lots of detail and feedback at each step. The idea is, we could tell developers to "first get your code working against this extra-friendly OAuth API, and then turn it to twitter or whoever else you want o hit, since then you'll be confident that it at least works in theory already." And yes, I agree that these types of developer aids are necessary for all the Open Stack building blocks!

Re: OAuth is working here! (Scripting News)

jsmarr — Tue, 17 Feb 2009 22:07:27 -0000

Dave, you inspired me to write up a list of ways we can make developing OAuth less painful: http://josephsmarr.com/2009/02/17/implementing-... -- now to actually implement each of these suggestions! :)

Re: Test-Driving the New Hybrid

jsmarr — Wed, 04 Feb 2009 14:53:01 -0000

Awesome Brian, thanks! You can't tell an OpenID story without JanRain--you guys are too good. ;)

Re: Introducing Two-Click Signup, an initiative to improve the user experience of OpenID; first test now live with Google

jsmarr — Mon, 02 Feb 2009 18:23:27 -0000

The Yahoo integration was also using an OpenID extension to share more user data during onboarding, but it was using the existing "simple registration extension" ("sreg") whereas with Google we're using an OAuth extension and then calling their API. Complicating matters further is we're also using the "Attribute Exchange" (AX) extension with Google to get some additional info, e.g. the user's validated e-mail address. AX is a generalized version of what sreg provides, and they're both complementary to providing an OAuth token, which involves more work but leads to greater flexibility in the data that can be shared. These are all different ways of basically doing the same thing, so over time things may sort themselves out more, but for now that's the way things are all working, and our general philosophy has been to just try and make progress along as many promising fronts as we can, so the market can sort out what ultimately works best. :)

Re: Introducing Two-Click Signup, an initiative to improve the user experience of OpenID; first test now live with Google

pauljacobson — Sat, 31 Jan 2009 09:06:58 -0000

Hi Joseph

I sent in a request via Identi.ca (for a change) and will keep reminding them. I have been pushing them to incorporate Creative Commons licensing support into Disqus so people can license their comments. I would like to be able to set a default CC license for comments posted on my sites in line with my terms of use.

On that note, what about CC license support in Pulse?

Re: Introducing Two-Click Signup, an initiative to improve the user experience of OpenID; first test now live with Google

jsmarr — Fri, 30 Jan 2009 17:23:27 -0000

Ugh, they do support OpenID in the sense that you can sign up for Disqus using OpenID and then login to leave your comment via Disqus, but they don't make it easy to do from their widget, which drives me crazy. I've mentioned it before but it hasn't risen to the top of their stack, so please help me provide "vocal user feedback" that this is a desired feature! :D