Disqus - Latest Comments for dacort

Re: Easily query AWS service logs using Amazon Athena

dacort — Fri, 06 Sep 2019 15:49:49 -0000

HI Michelle,

It looks like you you may not have provided the --extra-py-files parameter. Take a look at the "Before you begin" section - you'll need to copy a set of supporting libraries up to S3 to use as part of this job. Also Glue has released some new functionality since this post so the "Type" you want to select is "Spark" and the Glue version should be "Spark 2.2, Python 2". https://uploads.disquscdn.c...

I've included a sample job below (and corresponding image) that shows the different parameters:

{ "Job": { "Name": "cloudtrail_converter", "Description": "cloudtrail log converter", "Role": "AWSGlueServiceRoleDefault", "CreatedOn": 1567621625.803, "LastModifiedOn": 1567621625.803, "ExecutionProperty": { "MaxConcurrentRuns": 1 }, "Command": { "Name": "glueetl", "ScriptLocation": "s3://<your_bucket_here>/agsl/glue_scripts/sample_cloudtrail_job.py", "PythonVersion": "2" }, "DefaultArguments": { "--TempDir": "s3://<your_bucket_here>/tmp/glue", "--converted_database_name": "awslogs", "--converted_table_name": "cloudtrail_optimized", "--extra-py-files": "s3://<your_bucket_here>/agsl/glue_scripts/athena_glue_converter_latest.zip", "--job-bookmark-option": "job-bookmark-enable", "--raw_database_name": "awslogs", "--raw_table_name": "cloudtrail_raw", "--s3_converted_target": "s3://<your_bucket_here>/converted/cloudtrail/", "--s3_source_location": "s3://<your_cloudtrail_bucket>/cloudtrail/AWSLogs/<account_id>/CloudTrail/" }, "Connections": {}, "MaxRetries": 0, "AllocatedCapacity": 10, "Timeout": 2880, "MaxCapacity": 10.0 } }

Re: Seattle startup community mourns passing of Tom Music

dacort — Tue, 19 Apr 2011 13:20:35 -0000

I remember *somebody* filming the presentations on Sunday night, but can't recall who. Maybe Josh Maher?

Re: http://www.tweetdeck.com/twitter/Gillis57/~6eDHV

dacort — Fri, 01 Apr 2011 15:12:05 -0000

#golfclap

Re: The War For Talent

dacort — Sat, 19 Mar 2011 18:41:55 -0000

That's nothing. SEOmoz in Seattle is offering a $12k referral for both the referrer and the employee (http://www.seomoz.org/blog/.... And it's only getting more difficult with bay area companies opening development offices up here.

Re: Sharding and Replica Sets Illustrated

dacort — Mon, 15 Nov 2010 22:02:27 -0000

One of the reasons I was thinking of doing this is our load won't be evenly distributed due to an existing unique index on a serial value. One shard will be very write-heavy, while the others will have occasional reads scattered across them.

Your comment definitely makes sense, though, and I'll have to think through some more before making the final decision. Thanks!

Re: Sharding and Replica Sets Illustrated

dacort — Sun, 14 Nov 2010 15:36:39 -0000

How realistic is it to run a setup like this in production? I'll probably post this question to the mailing list, but I'm working on sharding a fairly large collection. My assumption was that I'd shard over two servers with two other servers as replica sets, but I'm wondering if this would be a decent approach. It would certainly allow me to keep more of my indexes in RAM than across two servers, but I'm concerned there would be some contention between the different "master/slave" processes. That said, due to sharding key restrictions, I'm going to be write-heavy on one shard with occasional reads scattered through the remaining ones.

Re: Help! As TeachStreet does better, I keep working harder and longer (and probably not smarter)

dacort — Sun, 07 Nov 2010 06:21:56 -0000

Wait, putting on weight isn't part of the startup life? Man, I've been misled!

Thanks for the reminder, mostly because it's good to know I'm not alone in putting every waking hour towards the startup. I've personally been trying to hit the sack before midnight but (obviously) not succeeding very well at that. 1am rolls around and I'm still cranking and for every hour I go to sleep after midnight is 2 hours past 7am that I wake up. The answer? Got me. The only thing that comes close is making sure we try to force those damn healthy aspects of life upon us. ;) I think Grant has a great idea that we can drive each other to be more healthy. I've been running more lately, let's try to get a schedule going!

Re: http://kylebanker.com/blog/2010/09/21/the-joy-of-mongodb-indexes/

dacort — Mon, 27 Sep 2010 18:33:54 -0000

You mention the following:

"An index on ingredient can and should be eliminated if you have a second index on ingredient→cuisine."

Is this still true if you have an index on cuisine→ingredient? I know it wasn't in pre-1.6, but the docs mention this for 1.6.x:

"Now you can also use the compound index to service any combination of equality (and some inequality) queries from the constitute fields."

I'm not entirely sure what that means, but it sounds like the index can be in any order?

Re: Track Social Media Campaigns from Inside Microsoft Excel

dacort — Fri, 13 Aug 2010 12:48:36 -0000

Hi Andrew - at the moment we only track Twitter and Facebook, but we're looking at the best way to expand that reach. We realize coverage is extremely important. The boolean setup is pretty slick and very easy to set up to require or exclude different words/phrases in the tweet. And while we haven't added sentiment, it's on the radar. We're still evaluating solutions that are more effective than some of the others that easily miss things like sarcasm and slang.

Sign up, take a look a look around, and we'll keep you posted!

Damon (@dacort) from RowFeeder

Re: Accepting Payments on the Real-Time Web

dacort — Sat, 05 Jun 2010 23:07:19 -0000

Hi Matthew, thanks for the kudos. It's much appreciated, we've put a lot more effort into RowFeeder lately.

As for billing - we ended up going with Recurly. Truth be told, though, were I to make the choice right now I would likely go with Chargify. They've been charging ahead (pun intended) with new features based on user feedback. Recurly is in a bit of a state of flux right now and it isn't clear what their pricing plan is going to be.

Re: Data for The Location Fight: Foursquare vs. Gowalla

dacort — Thu, 03 Jun 2010 13:46:13 -0000

Definitely - I know Gowalla had a pretty interesting deal with Adobe where they dropped "virtual" copies of CS5 in different locations that you could win when you checked in. I'm not sure how many other similar deals they've struck, though. The complexity of passports, and picking up/dropping off items with no real purpose was also confusing to me.

Re: Tweet Location Normalization in RowFeeder Enterprise

dacort — Thu, 13 May 2010 12:35:37 -0000

Hi Ellie - one thing to note is that even if you send FourSquare checkin to Twitter, you still have to have location enabled on your Twitter profile for the GPS coordinates to show up. That said, it's still possible to pull geo-coordinates by looking up the FourSquare venue.

Location is not enabled by default on Twitter, so adoption will be slow, but it's nice that they make it an opt-in.

Re: Accepting Payments on the Real-Time Web

dacort — Mon, 03 May 2010 12:02:20 -0000

Good to hear, and with 403 Labs as well. I know a few of the guys that work there, coincidentally enough - good group.

As far as the others, I believe you're correct in that they are still in the process of obtaining PCI certification. I know one of Recurly and Chargify were already compliant at a lower level and trying to get Level 1 certified, but I can't recall which at the moment.

Is Spreedly going to be on the list of validated service providers at some point? http://usa.visa.com/merchan...

Thanks for the update - it's been amazing to see a general lack of understanding around PCI in the startup world.

Re: Untitled Startup | RowFeeder

dacort — Tue, 27 Apr 2010 00:36:13 -0000

Additional @RowFeeder features we're currently considering/working on:

Export straight to Excel/HTTP/other, remove Google from the equation
Geo-normalizaiton/filtering of tweets (filtering exists, adding location)
Aggregate stat tracking
Custom per-client pre-processing

Re: Untitled Startup | Tweet2Blog

dacort — Mon, 26 Apr 2010 23:16:23 -0000

Shannon Whitley (@swhitley) has a tool to do this for your friends/followers, but not for lists (yet? ;)). More details here: http://www.voiceoftech.com/...

Re: Chirp Twitter Client Analysis

dacort — Tue, 20 Apr 2010 16:09:04 -0000

Just Apple's spreadsheet app, Numbers. Leave it up to them to make stuff look great.

By the way, if you want some of the raw #chirp/@chirp data to analyze yourself, just let me know. Happy to share - there's the spreadsheet I linked above, but that doesn't have app data.

Re: Untitled Startup | Don't Spoil It

dacort — Wed, 07 Apr 2010 17:21:56 -0000

Yea, it's unfortunate that Twhirl is no longer supported/being developed. That used to be my favorite Twitter client.

I wish more clients supported custom API endpoints like Tweetie 2 does. If that happened, I'd have more motivation to build up the little thing I cranked on to remove sxsw, foursquare, and gowalla tweets.

Re: The 5 Stages of Being a Twitter Developer

dacort — Sat, 20 Mar 2010 16:19:10 -0000

Hah, good point. This was made even worse that TweetStats originally scraped the Twitter.com website and Twitter had a tendency to make changes to the website on Friday nights.

So there'd I'd be at 5pm ready to head out of the office, and some major change would get deployed. Ah, good times.

Re: The 5 Stages of Being a Twitter Developer

dacort — Sat, 20 Mar 2010 16:06:47 -0000

Yup, definitely still counts! I wish I had realized sooner that I wanted to do this full-time. Definitely lost a lot of opportunity that way. Alas, I'm here now. :)

Re: The 5 Stages of Being a Twitter Developer

dacort — Wed, 17 Mar 2010 19:20:17 -0000

Hehe, I know. I'm a little disappointed by that a swell.

Re: Growing Pains

dacort — Tue, 16 Mar 2010 01:55:05 -0000

I know right where you are. I spent the past two years even just figuring out exactly what it is I want to be doing. It's been a long process of "do what you love and the rest will follow," but ultimately rewarding.

Re: Untitled Startup | Don't Spoil It

dacort — Mon, 15 Mar 2010 16:30:49 -0000

I actually whipped up something this weekend that was initially to filter out sxsw tweets (I'm a hater, I know), but support to remove the generic foursquare/gowalla checkin/mayor messages was quickly added.

Tweet 2 has a great feature where you can specify a custom endpoint and I took advantage of that. It's actually kind of nice having a clean stream. Unfortunately it's just a proof of concept at the moment and doesn't allow customization, but ... who knows! :)

Re: Untitled Startup | Where's Da Love?

dacort — Mon, 15 Mar 2010 16:16:36 -0000

I haven't done a sample lately to see how much larger geo has gotten on Twitter. It was initially a small fraction of a percentage, but I'm sure it's gone up especially since Twitter has started promoting it more.

That said, I had a friend that asked a few people. 12 out of 13 weren't going to enable geo functionality. There's always the location field, but the accuracy of that isn't the best either.

Re: Untitled Startup | Tweet 'n Save

dacort — Sat, 13 Mar 2010 17:58:05 -0000

Yea, I've been keeping an eye on the #crunch iPad promotion. Definitely pretty interesting, I wish I knew how they were tracking that.

Re: Twitter Enables Screen Name and User ID Lookup

dacort — Thu, 11 Mar 2010 13:08:35 -0000

True, it's easy to blow through that in ... a few seconds.