DISQUS

thanks mang

so has it hit the appstore yet or are you guys hacking on it now?

also: kenberger is totally spot on :-)

Everywhere you go in life there are bound to be risks, I'm sure you and your family are well aware of this... but let's recap:

- terrorist attacks could happen again (probabilistically you're more likely to get hit by a moving vehicle or mooing cow, though, while in India :-))

- india could go to war with Pakistan - if it happens, it'll happen in the north west near kashmir, that's where altercations would occur, or at least occur first - since it's currently not monsoon season in India, you can safely venture down as south as you like. Though war between India and Pakistan is possible, it's not exactly likely either (it would not be taken lightly by the international community)

- your kids are extremely fortunate to have this opportunity

- you are fortunate to have this opportunity

- you are going to India and not to a resort in the carribean for more than one reason - i'm sure this is not lost on your children

I would say that if your kids aren't scared or averse to going, you shouldn't be either.

Best,
b

Noah: relevance. Ads help subsidize a lot of the "free" good content we know and love - more relevance and value exchanged for small amounts of attention is one of the better things we can hope for.

btw, I tried to write a ruby lib to wrap the particular API you're talking about (if it's the same one I'm thinking) some time ago - I started it and gave up after a while (perhaps that's what you're looking at?) Back then there was a single encryption scheme, blowfish with ECB block cipher (and a non-obvious padding scheme - nul bytes :-( - non-obvious because one wouldn't think a nul-byte scheme would be used since it only works for non-binary data). ECB sucks. I was told that https was "too slow".

I spent some time in person speaking to the developer there, nice guy and all, but I walked away from that admittedly pretty spirit deprived.

Coincidentally we're trying to revive the lib here now, but are hitting problems with the crypto bits again -- I think something was inadvertently tweaked again on their end.

Anyway, if we're talking about the same lib, I believe they've got a way to enable "https mode" so that you can avoid the whole blowfish + ECB/CBC/whatever symmetric mess.

btw, I tried to write a ruby lib to wrap the particular API you're talking about (if it's the same one I'm thinking) some time ago - I started it and gave up after a while (perhaps that's what you're looking at?) Back then there was a single encryption scheme, blowfish with ECB block cipher (and a non-obvious padding scheme - nul bytes :-( - non-obvious because one wouldn't think a nul-byte scheme would be used since it only works for non-binary data). ECB sucks. I was told that https was "too slow".

I spent some time in person speaking to the developer there, nice guy and all, but I walked away from that admittedly pretty spirit deprived.

Coincidentally we're trying to revive the lib here now, but are hitting problems with the crypto bits again -- I think something was inadvertently tweaked again on their end.

Anyway, if we're talking about the same lib, I believe they've got a way to enable "https mode" so that you can avoid the whole blowfish + ECB/CBC/whatever symmetric mess.

I think I know which API you're talking about. :-)

I think I know which API you're talking about. :-)

My friend Mina put together something similar for our purposes, only he also has some controller magic so that you can easily refer to a polymorphic resource's parent from the polymorphic resource's controller.

I've pointed him at your post so hopefully he'll comment with details.

My friend Mina put together something similar for our purposes, only he also has some controller magic so that you can easily refer to a polymorphic resource's parent from the polymorphic resource's controller.

I've pointed him at your post so hopefully he'll comment with details.

My friend Mina put together something similar for our purposes, only he also has some controller magic so that you can easily refer to a polymorphic resource's parent from the polymorphic resource's controller.

I've pointed him at your post so hopefully he'll comment with details.

Peter, I think you missed out on Lijit's ability to overlay useful information in your Google search results, based on your network. Social effects as represented in the Lijit network ultimately affect the way you consume information. Attention-filtering with consideration of your social network works surprisingly well when consuming lots and lots of information (e.g., web browsing).

I think Lijit could do, and is in a position to do, many more useful things with the network data. Filtering, prioritizing, and sorting information can extend far beyond search (e.g., to the newsreader).

Scoble: That's true regarding the discovery of a parallel tech blogger concentration. Not sure how the dynamics of disparate but related clusters works, but I bet a huge time-evolving link graph of sets of bloggers would make for some really pretty eye candy. :-)

Scoble: I agree with what you said about linking to me, but I don't think you heard me out. My point was that TechMeme links to people who link among themselves and these people link because they also read each other (and so respond to each other). TechMeme seems to select memes out of a list of crawled feeds and that list is constantly updated, influenced by the feeds that were in it previously. This means that for me to STAY in the list, I have to be continuously grabbing your (or some other list member's) attention. For me to be continuously grabbing your (or some other list member's) attention, you need to be reading me; so I'd very likely be in your newsreader.

So then consider all the people in the list on Techmeme right _now_ and consider their RSS subscriptions. They likely allocate continuous attention to each other.

And this is all fine, really, and I'm not ragging on TechMeme because all it does is highlight this localized attention cluster, and it happens to be of some high profile web people, and that's great for traffic (for Gabe). That was my point. In fact I'm sure Gabe knows this; just consider his other web properties (focusing on attention clustering in the baseball fan + blogging subculture, as well as attention clustering in the tabloid + blogging subculture). There is a pattern of success here: he finds niches/accumulations of attention and builds a meme finder for them. Brilliant, no doubt.

What I was getting at is that there are lots of other attention clusters in the blogosphere, some smaller than others, and TechMeme is not likely (I think) to repeatedly highlight those. That's what makes it _Tech_Meme, after all.

I think TechMeme is really a selective memetracker optimized for page hits. And that's fine because presumably this is what Gabe wants.

I see TailRank as a better platform for doing meme tracking in a general sense because it allows you to influence the seeding, and allowing that is really key for people like myself who care.

But for the remainder (~80%?), TechMeme is good enough. It's good enough because the only person influencing how it's seeded (and what pops up at the front) is Gabe, and the people who get linked to are more often than not a closed-link community. I'd be curious to see the reading list/opml of every blogger linked to by TechMeme, intersect them all, and look at the resulting list of blogs. I bet they're pretty close to the blogs seeding techmeme.

I use Newsfire RSS for the Mac (http://www.newsfirerss.com/) exclusively these days, though I've seen Google Reader's latest incarnation and find it compelling, but Newsfire RSS lets me read my feeds offline -- not a must but really a must-have. There are only two features I'd really like to see in Newsfire RSS, but I'm not sure if they're best suited for a desktop RSS app.

Newsfire is 2-pane, with a feed listing on the left and the feed view on the right. I have my feeds organized in groups which I have to refine as my interests change, but currently they're something like this:

1. Basecamp Projects: RSS feeds related to Basecamp (www.basecamphq.com) projects.

2. Blogs (1st Tier): close friends and read with minimum latency.

3. Blogs (2nd Tier): bloggers I like to read when I have time to do so (they tend to write longer entries).

4. Blogs and News (3rd Tier)

5. Design

6. Architecture

7. Personal Content: feeds for my own content

8. Podcasts

9. vlogs

10. RubyOnRails

11. Social bookmarks (del.icio.us feeds of people I follow closely mostly)

12. Social Photos (Flickr photos of people I follow closely)

Now I usually read these in Newsfire's river-of-news way, which consists of pressing the spacebar to cycle between feeds. I usually have it set to "oldest to newest" river-of-news order for every new "set" of downloaded entries (i.e., every time I refresh and get new entries, i read oldest-to-newest for that newly refreshed set).

However, I can skip a particular feed in the river with a simple key combo (cmd + ctrl + down).

The two missing features I miss:

1. Ability to star particular entries and publish a feed of starred entries, possibly pushing starred items to del.icio.us (currently I click through to the entry and hit the 'del.icio.us' button in firefox to bookmark).

2. Ability to sync read state with an online server that I can then connect to from my mobile with a simple mobile RSS reader I can mark items read with.

I must admit I don't always agree with what Dave Winer writes but his take on the "podcast" trademark debate is spot on, and I think it's fair to Apple.

As for my favorite podcast: Coverville! http://www.coverville.com/

This looks like exactly what I was looking for, almost. In particular this is cited as a feature: "Store, record and burn TV content onto DVDs using HAVA, enabling the PC to operate as a Personal Video Recorder."

Now if only it worked on Windows _and_ OS X, I'd buy it.

1. Regarding old habits being tough to break: eventually you'll probably realize that the value you get from changing is greater than the cost of changing and you'll change. It happened to me with my real official switch to mac, and I'm sure it'll happen to you.

2. Regarding syncing with offline: on the contrary, gcalendar syncs with iCal (if you're trying to figure out how to pull your calendar data out take a look at this: http://www.google.com/support/calendar/bin/answ...) iCal is great for everything I've ever needed to do offline. Oh, and I have a PocketPC/smartphone, and there is this neat little ActiveSync replacement for the mac called "PocketMac Pro" (here's a great review: http://maczealots.com/reviews/pocketmac/). Why do I mention this? Because it syncs iCal with the calendar in my PocketPC (it also does iPhoto, address book, and all the other goodies!)

3. Email + Calendar integration: this is getting better with gmail + gcalendar by the month. Take a look at this post for some of the details: http://reblogger.wordpress.com/2006/04/24/gcale...

4. enterprise contact management -- gmail remembers who you sent stuff to, and who sent stuff to you. it remembers that, _and_ context, so when you pull up an old email, you get the entire thread right there and then. I don't know about you, but I'm subscribed to a whole bunch of mailing lists and I extensively use gmail's filtering and labels features. It's gotten to a point where one of my "actions" for certain classes of work Email (that I get via Thunderbird+IMAP) is to forward to my gmail account at myname+some_special_action_label@gmail.com -- I've learned to love this feature as it's really changed the way I work, for the better.

5. seeing other people's calendar use case: I'm not completely sure on the details here, but I have "intersection" calendars set up with people I meet with often. Admittedly I haven't used them much (yet) -- in both the case of gcalendar and exchange, both people have to be using the system, but at least gcalendar is more platform agnostic.

cheers,
bosko

I use NewsFire on the Mac and a multi-tiered approach: I have several categories (Tier One Blogs, Tier Two Blogs, Tier Three Blogs, Podcasts, News, Social Bookmarks, Social Photos, My Content) and I sometimes cherry-pick on categories I read depending on time. News is usually the most noisy and least relevant as it is various sources from NYT and other high-volume professional publications.

I have NewsFire set up so that when I click on links it magically opens up new tabs in Firefox. I then cmd+tab over to delve deeper when I want to.

I think it sucks that after several years of doing this stuff we're still stuck with the same damn problem of having to constantly context switch from one application to another to consume a large inter-related set of data. Feed consumption needs a new usability/UI paradigm, badly, and it needs it right now. Unfortunately, much easier said than done.

I find I "discover" a lot less when it comes to podcasts (as opposed to blogs or other text-content-driven web media) largely because it is very time consuming to just 'quickly scan through them and pull out the good from the bad, or vice-versa.'

I wish there was not only a directory, but one where ranking was truly community-driven and reliable, and where podcasts were at least meaningfully categorized. Odeo is probably the closest thing out there to this right now, but it's not perfect.

It's particularly ironic that humans need simple top-of-category style directories, to make it easier for them to consume "good content," they in part need to be told what "good content" is. If this happens in a community-driven way, I expect that roughly 90% of listeners will listen to some top 10% of content, leaving the long tail collecting dust.

Scoble, it's awesome that you guys decided to do this.

I blogged about how I thought it would be awesome if you did a while back:
http://www.crowdedweb.com/articles/2006/04/24/c...

And you did it. Kudos!

-b

I just read through the remainder of the comments pertaining to the fact that the current spreadsheet market is largely dominated by enterprise. While a lot of these comments are spot on, they neglect to note that just because the current market appears to be dominated by enterprise does not imply that there is no other market for spreadsheets, just that the current model is not suitable for that other market.

A mediaware model where basic spreadsheet functionality is free (and perhaps subsidized by relevant advertising, lead generation, or similar) _is_ more appealing than a relatively expensive and over-featured piece of software that does too much for non-enterprise spreadsheet users. And these users exist. They simply don't dominate the _current_ spreadsheet market... for now anyway.

Scoble, if you compare MS Office to Google "Office" (to be?), you _have_ to compare them in terms of cost/value w.r.t. the end-user.

What I mean is this: Google Spreadsheet might very well be very under-featured compared to MS Excel in MS Office 2007, but it's also free. Collaboration is easy in GSpreadsheet in large part _because_ it's free and easily accessible/adoptable. If you want to start comparing thin-client with thick-client, it would help if both sides offered the same value proposition. But the fact of the matter is that right now, MS Excel costs money (i.e., is software), while GSpreadsheet is media-ware (free for basic use but potentially monetizable attention).

I think that if MSFT is really serious about the thick-client approach, then tie it in to web services (with enough hosted intelligence to be able to leverage it in the form of mediaware) and distribute the thick client for free _now_ and be ahead of the curve. If what you're predicting is true, and Google does have to "thicken" the browser to have GSpreadsheet do the same neat things Excel already does, you can bet your bottom dollar that when they do, it'll be free, and at that point the Excel featureset will already have become commodity.

You can still charge for the Office _suite_ and perhaps offer some additional membership services on top of it, but stop distributing your software like software (using the old model) if you're really trying to become a media company.

Cheers,
Bosko
http://www.crowdedweb.com/

I like Channel10. I also enjoyed the Channel9 interview. Robert, I'm wondering, could you produce an interview with the people who choose and produce the content on 10? It seems that there is a dual value in 10, the first being the community side of it, and the second being a potential content consumer community (presumably they will overlap). It would be interesting to learn more about who drives the content creation + production at 10.