HN user
email: thom@ below domain
work: https://telcom.uk
Yeah it's certainly not without issue, the network is full of blockages, collapsed ducts etc.
Traffic management and road closures can be hard work, we've had to wait over a year before for a road closure as it would affect multiple bus routes. (And as an aside, lockdown was extremely productive for network build like this!)
For clarity - whilst most connectivity infrastructure in London is underground, it's almost always within a primary duct, so running new infrastructure is usually a case of pulling in a new cable as opposed to "ripping up the street".
In fact, anyone approved can use BTs own ducts and poles via their PIA product[1], which has created a resurgent and incredibly active market of "alternative" network providers ("alt nets"). London for example is now well served for broadband by Community Fibre, g.network, Hyperoptic and others alongside the incumbents.
[1] https://www.openreach.co.uk/cpportal/products/passive-produc...
This looks to be rewrapped version of the open source knowledge base Outline: https://www.getoutline.com / https://github.com/outline/outline
Specifically, their editor is a separate component: https://github.com/outline/rich-markdown-editor
I assumed this was a clever marketing product from the outline team, but it seems to be entirely separate.
I'm planning to release this as a plugin :)
Recent TransferWise horror story of a public company, processing millions, who had their account suddenly shutdown: https://blog.simwood.com/2020/03/emergency-bank-details-chan...
Both are valid addresses and can be assigned to hosts.
Network address is only really used for directly attached networks, non directly attached networks will route to any address in the block correctly.
Same for broadcast address, they're also relative to whatever block you're talking about at the time, so whilst 3.255.255.255 is the broadcast for 3.0.0.0/8 subnet it's just another "usable" address in the 3.0.0.0/5 subnet and when you send a packet then you, and probably your router, don't know what subnets in use on the other side :) (unless it's directly attached)
Too many customer devices don't support /31 subnets unfortunately, for example with Draytek we've seen an issue where it would accept the 255.255.255.254 subnet but we'd see a whole raft of connection issues making the connection unusable.
If we provide a dedicated IP for a connection where we provide the LAN we just put a single /32 on the loopback and NAT onto this which is obviously much more economic with addresses.
We're an ISP, most of our customers are businesses. Of those, around 50% opt for a pre-configured LAN (i.e. we do NAT and usually CGNAT too).
For the rest we provide a static IP address, so we'll allocate a /30 (block of 4), and they get a single usable address which they will assign to their own manged router/firewall.
For the majority of our customers "networking" is either handled as overflow for their in/out IT resource or often by someone remotely savvy with tech.
For most of these people networking ranges from an infrequent concern to a vague mystery that can be sorted with a bit of googling.
For most, deploying and testing IPv6 has absolutely no upside and quite a bit of potential downside, that's because "everything works" on IPv4 and configuring IPv6 is just another potential source of error.
In addition, most people who opt for this setup do so in order to expose some internal service to the internet (port forward), again there is usually zero incentive to also deploy IPv6 as they can't be sure their client device will be using v6 when they come to connect, but they know it will support v4, everything does.
And so herin lies the issue, it's chicken and egg, they know they need v4, not every server they access or client who accesses their forwards supports v6, so they _have_ to implement v4. As such they see no reason to "faff" about with IPv6, and I don't really blame them.
We're considering charging more for dedicated v4 and possibly offering a free translation service (another point of failure :() but honestly, most would just pay the extra and then just resent us a little more. Our competitors continue to acquire v4 space as we do, this is what our customers want.
Until there is v6 only content (but who is incentivised to do this?) then I can't see any incentive for these users.
Homo Deus (the sequel as mentioned) answers the question of why he has formed his particular view of the future.
In essence, yes economic incentive is the basis for much of today's human corporation, but with growing AI/robotics, the same economic incentive will make much current human input unnecessary.
Then, for example, would counties see so many humans as a big cost? How do they solve that if they no longer need them to work?
Scary stuff.
We took a slightly different approach to solving a similar problem: https://github.com/thomseddon/traefik-forward-auth
We were already using traefik as a proxy for our docker/swarm clusters and this is a single container drop in to add authentication to every traefik request.
It's still missing a few key features but it can get you started, we're testing the use of a single auth domain (so you don't have to add every internal service domain as a refirect_uri in Google - looks similar to how sso works) internally and we expect to release this shortly once finished.
Additionally, if you want an even lighter weight option, we also use, with great success, cloudflare's lua script on a few services we don't run with docker/traefik: https://github.com/cloudflare/nginx-google-oauth
I think it's safe to say you're underestimating Disney, they don't just knock out a kids film every year or so - they invest heavily in R&D and are pioneering many tech fields, animation, haptic feedback etc.
They know how to do tech, ever heard of the magic band? https://www.wired.com/2015/03/disney-magicband/
Found myself in the exact same situation (have a WileyFox, updated and ended up with the awful true caller app as the default dialler + found I couldn't use the default google dialler), I've since been using: https://play.google.com/store/apps/details?id=com.contapps.a... but it does have adverts, I'd welcome any other suggestions (aside from re-flash).
Only today I was joking about how absurd it is that I was struggling with such a fundamental feature.
Looks like you can get 100Gb NICs for servers e.g. https://www.hpe.com/uk/en/product-catalog/servers/server-ada... (Intel, Mellanox etc also seem to have their own)
On the distance, there are other specs that support 100Gb up to 20km on there!
If you're wondering what this kind of optic costs the rest of us: http://www.fs.com/products/65219.html (£1,081)
Although you'd pay at least 10-20x that if it had a big vendor name on the top.
Chris, after starting an ISP[1] in the UK in 2014 I certainly feel you're pain around the difficulties in starting an ISP! Myself and my co-founders have often looked at the comparable easy of growth software based startups and wondered what on earth we were doing, but we soldier on and our looking at our reliable, contracted MRR now is certainly starting to make up for it :)
Good luck, I look forward to keeping up with how things are going your side of the Atlantic.
Thanks for writing the post, I built a moderately popular extension in 2012 "Twitcher" for switching between twitter accounts (https://chrome.google.com/webstore/detail/twitcher-twitter-a...).
I've toyed with different monetisation strategies over the last couple of years, I even conducted a 6 month survey of what people would be willing to pay. In the end I decided I couldn't justify charging for it and a few months ago I rolled out a "pay-what-you-want" model like AdBlockPlus, during the A/B test on 20% of users I generated just over $100 dollars in the first 2 weeks, then basically nothing - so the annual/monthly model is attractive!
I like the idea of keeping it open source, then those who don't want to pay can still use it - think I may follow your model :)
They usually do, but these only protect from grid failures. HEX certainly does have battery + generators (it's one of the most important DC's in the country).
Components inside the DC should be redundant, but the customer can decide whether to have dual power feeds from redundant areas of the DC provided to their rack/cage.
Even with all this, issues do happen, Telecity had a very high profile and prolonged issue at the end of last year that affected AWS & Azure amongst others: http://www.theregister.co.uk/2015/11/18/telecity_outage_fix_...
It's a power outage in Telecity Harbour Exchange, from the LINX mailing list this morning:
08:12
- We are currently experiencing an outage impacting some of our equipment located at Telecity/Equinix Harbour Exchange in London. It is impacting both the Juniper and Extreme LANs.
- We are still investigating to get further information, but we have lost all management access to the devices at the moment.
08:45
- We have had confirmation from Telecity/Equinix that there was indeed a partial loss of power impacting one of the power feeds.
- Power failed at 07:55 BST and was restored at 08:17 BST for the majority of our equipment.
- Unfortunately one of our edge routers (edge4-tch) on the Juniper LAN is still down at the moment.
09:33
- Power to edge4-tch has also been restored now. The router has been back up since 9:15 BST.
- All member services should be restored.
- We have an engineer on site and are working with Telecity/Equinix to understand why we lost all power to that device.
- We are also waiting for a full report from Telecity/Equinix about the cause of the power failure.
---
We don't peer with BT so ou traffic was OK but we did notice loss of service from AWS to BTNet
Looks great! I also find the other way around very convenient: http://thomseddon.github.io/cronstring/
Reddit use Cloudflare on both reddit.com and redditstatic.com
Why would they choose to host this on Wordpress VIP?
SSL.
That is all.
Yeah, because the only cost in broadband delivery is the undersea fibre??
How about the core network? High capacity core routers? Firewalls? Datacentres? Power? Optics? Engineering? Ongoing upkeep? Monitoring? Upstream bandwidth? Interconnects? Peering LAN membership? LIR membership?
In many ways their naivety isn't too surprising, we started an ISP last year (https://telcom.io) and the shear scale of the undertaking and costs baffels us everyday!
He did (stripe): https://remysharp.com/images/jsbin-stripe-fraud.png
We don't actually supply their infrastructure or offices, we provide connectivity for one of their co-working spaces :)
There really aren't many freely available DNS servers that have seen serious production usage - I'm really excited to see RRDNS open sourced as previously hinted at, does anyone know if there is a timeline for this?
Some say that this is all Google are really trying to achieve with Google Fiber...
Don't forget that on the consumer side, anything above 1Gb is an entirely different kettle of fish. We can achieve 1Gb over the most common cat5/e cables (1000BASE-T) but they simply don't support carrying anything more, same goes for the NIC on most computers, it's just not common to get anything above 1Gb.
The main benefit that is reasonably achievable for most people would be that they could aggregate at over a Gb, so they could connect multiple devices and expect a full Gb for each. Even then you would have to ensure they either both connect to the router directly or you have > 1Gb towards any switches (either 10Gb or LAG).
Achieving more than 1Gb on a single device would require the use of both cat6e cable or fiber and having a 10Gb NIC on that device - this is assuming the Comcast router even has 10Gb ports! And even then I'm sure they're would be some other hardware/firmware/software issues to tackle...
An interesting problem though....as for delivery, it could be using GPON, most active point to point is either 1 or 10Gb whereas most existing GPON ports support up to 2.5Gb which seems to fit better.
"$10K GBP"
Confusing