Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
kazen44
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
18 ms
·
181.
▲
by
kazen44
4y ago
this alienation is nothing new, and has happened during the industrial revolution and subsequent rise of things like taylorism. There is a reason it is a main tenant in karl marx his writings[0] Ofcourse people are not going to go beyond an
182.
▲
by
kazen44
4y ago
heck, when running something in production with multiple underlying servers, you need a system that builds an overlay network regardless if you use containers or Vm's..
183.
▲
by
kazen44
4y ago
ASlso, docker to me seems quite leaky in terms of abstractions (especially the networking side) compared to something like a freeBSD jail. why does docker networking need to be infinitely complex? freebsd solved this issue far more cleanly
184.
▲
by
kazen44
4y ago
this law is also in effect in a lot of other european countries. Remember that after World War One and during the interbellum, most european countries had massive social unrest. Giving workers representation inside the company was usually m
185.
▲
by
kazen44
4y ago
to add to this, a ton of houses are not build for heatpump based systems. Most radiator style heating (aka, anything build prior to the 2000's) will not work with a heat pump because the water temperature cannot get high enough to get
186.
▲
by
kazen44
4y ago
heck, Henry ford started this buy increases wages so much his employees where able to buy the product they themselves have build.
187.
▲
by
kazen44
4y ago
what about a company building housing for its employees. This is something that happened a lot in the netherlands in the city of eindhoven. Phillips had its major manufacturing located in the city, and build massive housing projects to hous
188.
▲
by
kazen44
4y ago
HW is also far more capital intensive to build compared to software, so usually far less companies have to resources to build HW based solutions.
189.
▲
by
kazen44
4y ago
starting a networking company (except consulting) is extremely hard. Building an ISP is even harder and requires insane amounts of money. Most of my friends in the network engineering field either work at companies that are large/compl
190.
▲
by
kazen44
4y ago
same goes for many other places in europe aswell. (netherlands, germany, france etc). just look at football rivalries to get the idea.
191.
▲
by
kazen44
4y ago
> the Sten had magazine feed problems the entire war because of contractors going off spec and the Soviets had innumerable problems with QA. If you can't trust outside firms then you have to build everything in house) Another good t
192.
▲
by
kazen44
4y ago
packet based networking was invented in the US... The internet was created by meshing many different networks together (mostly ARPANET and SFNET). Networks which now form the backbone of the internet also existed in europe at the time, main
193.
▲
by
kazen44
4y ago
> Really? In some places in Europe, people were starting to get excited about dial-up BBSes in the mid nineties. What is your point exactly? Half of europe was still transitioning to a market economy or in a (civil) war/conflict i
194.
▲
by
kazen44
4y ago
in my opinion, most european companies are not using the cloud (especially if they are not in the tech space). Colocated hosting is very, very large in europe, and many small/medium bussinesses operate out of a couple of VM's on a
195.
▲
by
kazen44
4y ago
it gets even more fun if you think about what people are actually paying for. For instance, do you charge for the overhead of the Ethernet frame? What about additional size thanks to tunneling/encapsulation? One could argue that no, yo
196.
▲
by
kazen44
4y ago
also, what do you count as bandwith? Simply sending packets down a wire usually limits one to the physical speed of the interface, but even then some kind of packets might get dropped by various filters/mechanisms along the way. For in
197.
▲
by
kazen44
4y ago
that is the problem with oversubscribing though. making it possible to say "you have a guaranteed bandwith of X% of the total capacity" is a hard problem to solve, especially inside the core layers of the network, where traffic ge
198.
▲
by
kazen44
4y ago
having resonable response times make a huge difference in being oncall. in my experience, most employer seem to think responding within 30minutes or so is doable, and i tend to agree with them. Usually makes it possible for you to just do t
199.
▲
by
kazen44
4y ago
the chance of two SSD's failing at the same time under normal circumstances is extremely slim. So this might actually be a good cause of this incident.
200.
▲
by
kazen44
4y ago
This is the case for many other european countries aswell. In the netherlands for instance, uber and many others got slapped down hard for circumventing the law according to the literal implementation of the law, instead of taking into acco
201.
▲
by
kazen44
4y ago
Heck, there is an entire spectrum of politics which state that pocketing excess value from the productions of others is wage-theft and thus exploitative.
202.
▲
by
kazen44
4y ago
what does this have to do with OP's post? Also, your question is far more complex and not comparable to the russo-ukraine war. For one, the geo-political situation prior to WW2 and especially compared to today might aswell be from a co
203.
▲
by
kazen44
4y ago
"punting it to software" from a router with seperate control and forwarding planes perspective, is forwarding it to a control plane, instead of relying on the logic programmed inside the ASIC to forward traffic.
204.
▲
by
kazen44
4y ago
how are you going to keep this counter? Do you identify the bytes that are processed in individual flows? Which system will keep track of this? the control plane of the router maybe? great... you just added additional complexity instead of
205.
▲
by
kazen44
4y ago
> I’m assuming it’s somewhat justified. in a sense, yes. People claiming software based solutions can match performance of hardware basic ASIC's are simply not thinking about the scale and speeds of modern core routers and switches.
206.
▲
by
kazen44
4y ago
compared to rolling out IPv6? definitely, especially on the longer term. For instance, most Core/Edge routers (my experience is mainly with juniper MX series, but i assume the model is roughly the same for other vendors), you need spec
207.
▲
by
kazen44
4y ago
also, ipv4 address space is becoming intrenched by large players because address pricing is becoming very expensive. Want to start a new service and require global connectivity, good luck winning IPv4 auctions because AWS is buying up all a
208.
▲
by
kazen44
4y ago
Yes, there could be. But in the end, those are just parameters of the specific networks/services that make up the internet. Considering that routing is not deterministic, you could get vastly different characteristics from point A to p
209.
▲
by
kazen44
4y ago
because maintaining state for GCNAT tables is far more complex then just forwarding packets. routers doing NAT are thus more expensive then those just doing simple forwarding. Also, in some countries ISP's need to map the use of a spec
210.
▲
by
kazen44
4y ago
most ISP tier routers have an entire commit engine to load and apply configs. junipers allows for instance, one to do the command commit confirmed, which will apply the configuration, and revert back to the previous version if one does not
More ›