GeneticGenesis

Author here.

You should have read the first draft

Wow what a terrible website. The GDPR advertising opt-out:

1) Renders at the top of the page when the page is loaded, then because this link contains an anchor scrolls you to that anchor, while the entire screen is greyed out

2) Your behavior is actively being tracked by default, every click, mouse move is sent to a third party.

3) The GDPR opt out process fails every time.

Solid violations of GDPR, well done.

Unrelated to the content, but the stock photos throughout this article felt completely needless to me. We're not 5 year olds, we don't need pictures unless they add to the content.

Anecdotally, 3M seem to suffer more than I'd expect from fake products flooding the market place.

I've lost count of the number of reviews of products I've read which advertise "3M tape / adhesive" or alike, where it turns out to be something cheap with a 3M logo on it.

Likely a drop in the ocean of 3M's business, but every little helps, I guess.

Interesting, thanks for sharing! Some questions:

Where is the data sourced from? Is there an offline DB? How often is the BD updated? How is reputation calculated?

It's interesting that no-one mentioned Formula 1 in here.

F1 was one of the biggest industries to embrace tobacco advertising, and indeed was exempt from several tobacco advertising laws in the UK. The Ferrari F1 team still takes sponsorship from Philip Morris International (the company that owns Marlboro), and indeed at the start of this year, the Ferrari team was named "Mission Winnow Ferrari". Mission Winnow is Philip Morris' project looking for Tobacco alternatives. Read: Vaping.

As @satori99 mentioned however, Australia has very strong advertising laws when it comes to tobacco, so as the first race is in a few weeks, the Mission Winnow branding was removed from the Ferrari as it was revealed that the Australian authorities had launched an investigation: https://www.autosport.com/f1/news/141908/ferrari-removes-mis...

Author here!

We've been playing with representing our APIs using tools like Swagger for a while and finally built a toolchain around OpenAPI V3. It wasn't all smooth sailing though, the tools aren't perfect yet.

Let me know if you have any questions or tips for tools or best practises we missed!

I've always found the HFT space interesting, but don't really have any desire to move into the financial sector.

How large are the payloads in general for HFT?

I disagree. I've seen P2P outperform traditional CDNs many times, take a look at some of the data from Peer5 I linked below around improving the user experience.

P2P doesn't need "always on" boxes in large scale live stream scenarios, which is one of the use cases where it works best. I think we'll see a lot of growth in hybrid P2P/traditional CDN in the next couple of years.

Peer5 is actually a really interesting example of how the user experience can be improved using peer-to-peer, particularly in areas with less capable internet infrastructures - their technology _only_ kicks in if the user experience can be improved by using peer-to-peer, and it'll fallback to traditional CDNs if they can't improve the experience.

Despite this logic, they were capable of very impressive P2P offloads during the world cup. Their datasheet linked at the bottom of this blog post is worth a read for more details: https://blog.peer5.com/what-a-world-cup/ (email address needed)

Great point!

The BBC actually did some really interesting work last year on using MPEG-DASH with Multicast with a demo at IBC, the research can be found here: https://www.bbc.co.uk/rd/projects/dynamic-adaptive-streaming...

As far as I'm aware there aren't any public CDNs that support the sub-200ms approaches which don't involve buying a full solution from that vendor.

* Limelight's solution is only really accessible through their low latency streaming products provided by Red5.

* Fastly and Akamai are both chasing Ultra low latency through chunked CMAF delivery.

On a slightly smaller scale, Cloudflare supports websockets, so you could use a protocol like the one Wowza are using (WOWZ). https://blog.cloudflare.com/cloudflare-now-supports-websocke...

If you're interested in a toolkit to build out a webRTC style CDN edge, I think the best place to start would probably be with Gstreamer's new WebRTC tooling https://opensource.com/article/19/1/gstreamer

My gut instinct is there's both technical and business reasons.

From a technical perspective, traditional CDNs are struggling to grasp what it means to be a CDN in the peer-to-peer space, their networks have been built to support traditional HTTP traffic, and they've been fairly successful delivering that. Adapting to cacheable peer-to-peer objects would mean large architectural changes.

From a business perspective I suspect they also see a challenge where encouraging people to chase peer-to-peer oriented solutions may increase adoption, and start to reduce the amount of traffic they're serving from their private edge as more viewers start to peer between each other.

Author here, there's lots of interesting innovations happening in this space. If anyone knows of anything interesting I've missed, please let me know!

I thoroughly encourage everyone to learn more about Tarrare on the This Paranormal Life podcast!

https://soundcloud.com/thisparanormallife/044-tarrare-the-ma...

Hey, your website seems to be down and there isn't really any documentation on your project - what sort of codecs, packaging and delivery formats will you be supporting?

Thanks!

Author here in the comments if anyone has any questions!

Author here! Let me know if you have any questions!

That's not really innovative - other CDNs are already doing the fast first few segments and then slower/cheaper CDN once the buffer is built up. Hola CDN for example. https://holacdn.com/multicdn

Dynamic Manifest generation has been done for years, first by Unified Streaming, now by most companies.

Looks like this is a Bitmovin resell - Player is Bitmovin and the DASH manifests are clearly created by them too. Storage appears to be Google Cloud storage with Cloudflare's CDN tacked on in front. Doesn't look like any radical innovation to me.

Sure. Whenever a "config change" (Note: this includes adding or removing targets to a target group, EG Autoscaling) happens on an ALB, the ALB drops all active connections, and re-establishes them at once, at high load, this obviously causes significant load spikes on any underlying service.

You can see this happening by looking at the "Active Connection Count" graphs from your ALB, and adding or removing an instance from an ASG.

At 30+GBPS and over 20kRPS, removing one instance can cause absolute chaos.

No chance that I'll jump into another new load balancer product from Amazon any time soon. ALB has significant deficiencies that AWS don't warn you about, and you only find then at tens of thousands of RPS.

Still waiting on that fix, AWS.

Have you filed a radar bug?

I want to have a deep understanding of Kubernetes...

Oh, and Finnish.

I do this for my family these days, virtually zero maintainance.

Biggest word of warning though is that Google are totally happy to change things with no warning or how or why, which can be very confusing.

For example, they stopped a direct route to import photos from an SD card or alike to google photos. No automated replacement workflow.

Another big blocker is cost.

For example, one of the biggest CDNs in the world (I'm looking at you, Akamai), charge dramatically more for delivering content over HTTPS. Let's say you're delivering video content at scale, the difference between HTTP and HTTPS delivery can be many millions of dollars a year.

But why not use a different CDN, say Cloudfront which prices the same for HTTP and HTTPS?, well, simple, the same problem. Cloudfront is many times more expensive than other CDNs at scale.

Really, we need to apply pressure to all CDNs to equal out their HTTPS pricing (Its not just Akamai...)