To be fair, this is exactly how I use Netflix. I have a free app that I pay to use completely outside the app store. Not sure of the actual rules, but it's definitely possible.
I'm really interested in this market. Does anyone know where having easy-to-build floorplans are actually useful? Especially without expensive HW setups like this?
Unless you are doing live video, there is no reason to transcode real-time. Do it all upfront. You need to decide on your requirements and transcode everything to your desired format.
From the research I've done, I'd suggest MPEG-DASH for reasonable cross-ish platform adaptive streaming without requiring fancy video servers. Your requirements and platforms might be different than mine though. Research is required.
This would normally be the case, but for small projects with a small number of contributors, they are always free to dual-license their own code, even "retro-actively". Anything that was already released would still be available under GPL, but new edits could in theory be licensed under whatever terms they want, independent of how "derivative" it is. For large open source projects without a copyright assignment clause, this is essentially impossible, but for small projects it's relatively easy, assuming you can track down all the contributors. IANAL.
Google is actually pretty good at accepting updates to things like operating hours. I've done it many times. They actually have a program (https://www.google.com/local/guides/) that encourages you to keep their maps up-to-date with correct information.
no. name picking is a rabbit hole, proving your product/service is far more important. pick something unique so that people can google it and come up with a meaningful result. If it becomes a problem, you can always change it AFTER you've proven your idea.
sorry man, it's just not that useful in it's current form. Ranking reviews by usefulness might help, but it's a babystep. I think what I would find more useful is being able to compare camera's and lenses side by side, although there are certainly sites that already do that. I think people care about 1) budget 2) lens ecosystem considerations 3) the "gotchas" (hard to quantify, but you see stuff like "this camera requires you to hunt through 7 menus just to change this one setting", or this would be a great camera except _____), which your site does very little to address.
Didn't get it. Found your ig profile. tried it. seems very un-intuitive. not sure people will understand it or go through all the trouble. [See Image -> goto profile -> click link in profile -> find image again -> click image -> go to website]
I feel like something that simply 302'd the user to the most recently posted URL would make more intuitive sense, but with obvious drawbacks.
To me the obvious use case is querying your log files as stored on s3. Query for a specific combination of features, or do some (simple) processing on them.
It's really only useful for a small list of file formats. Doesn't really do much for you if you primarily use s3 for binary data or static web hosting.
Anybody know whether it is kosher to use Rekognition (or comparable Google/Azure/Clarify solutions) as a source of "supervised" data to bootstrap your own in-house ML models? I'm pretty sure the TOS would frown upon that, but it would beat manually tagging your own datasets, plus its hard/impossible for them to enforce.
EDIT: in retrospect, the pricing means its probably not economically viable.
What's your usecase for Athena (Really curious how this changes anything)?
You can hook your lambda up to a cloudwatch scheduled event (http://docs.aws.amazon.com/AmazonCloudWatch/latest/events/Sc...). If you don't do much when this happens (like, you exit immediately), then you won't be charged much compute time. Can't find the reference to how often you should ping it to keep it warm, probably varies, IIRC hourly would be plenty.
Re: users auth. Isn't that what Cognito is supposed to be? I mean, I don't fully understand it, but I think so.
As for the cold boot issue, I thought the standing solution was to have a "fast-exit" ping-like code-path within the lambda. Query it on a regular basis (you can even do it with a lambda scheduled-event). That way your lambda should be kept warm.
Note the 10MB minimum "charge" per Query. For small datasets under 10MB, you'd only get up to 200 Queries for the minimum billable $0.01. That would be a fairly small number of queries, so probably not that useful. Plus you'd have all kinds of issues regarding consistency if your data was dynamic (s3 is a blob store, not a database, normal s3 consistency guarantees still apply).
I'm confused though. The monthly fees for dynamodb only apply after you exceed the free tier, and for someone who is unable to commit to a monthly fee because they envision low usage, shouldn't the free tier be sufficient? (Honest question, I'm looking at using dynamodb, but comments like this make me think I'm missing something)
"To help, recently Snopes has begun to publish important news-related information as a resource for journalists and others. Last week, for example, Binkowski wrote a piece about how the electoral college functions that served as a reference for other reporters writing about the election. “It’s really complex stuff that you can’t just read about and then write about,” she says. “We’re going to be doing more of that.”"
I did exactly this to in order to use cloudfront signed cookies with lambda/apigateway. API gateway only allows you to have one set-cookie header per response, but 3 are required for cloudfront signed cookies. (See thread: https://forums.aws.amazon.com/thread.jspa?threadID=205782)
UPDATE: Looks like there is a work-around that uses the fact that the set-cookie header is treated as case-insensitive by the browser but not apigateway. Genius.
Can't use it at all. Locks up with a redirect loop on registration. Cleared my cookies and tried logging in, currently getting 302's from home->setup-wizard->home->set-wizard->...