On Sun, 29 May 2022 07:44:51 -0000 (UTC), Muttley@dastardlyhq.com wrote:

>On Sat, 28 May 2022 18:01:30 +0200
>Rolf Mantel <news@hartig-mantel.de> wrote:

>>Cloud servers enabled a REST-ful programming paradigm, without which 90%
>>of our internet-enabled mobile phone apps would not be feasible.
>
>Yes they would. Apps are just programs, they have little to do with the web
>(steve jobs made sure of that) unless they're written in javascript.

How do you think mobile apps communicate with their data source, then?

Mark

On Mon, 30 May 2022 20:25:03 +0100
Mark Goodge <usenet@listmail.good-stuff.co.uk> wrote:
>On Sun, 29 May 2022 07:44:51 -0000 (UTC), Muttley@dastardlyhq.com wrote:
>
>>On Sat, 28 May 2022 18:01:30 +0200
>>Rolf Mantel <news@hartig-mantel.de> wrote:
>
>>>Cloud servers enabled a REST-ful programming paradigm, without which 90%
>>>of our internet-enabled mobile phone apps would not be feasible.
>>
>>Yes they would. Apps are just programs, they have little to do with the web
>>(steve jobs made sure of that) unless they're written in javascript.
>
>How do you think mobile apps communicate with their data source, then?

TCP. Next.

On Mon, 30 May 2022 19:38:26 -0000 (UTC), muttley@dastardlyhq.com wrote:

>On Mon, 30 May 2022 20:25:03 +0100
>Mark Goodge <usenet@listmail.good-stuff.co.uk> wrote:
>>On Sun, 29 May 2022 07:44:51 -0000 (UTC), Muttley@dastardlyhq.com wrote:
>>
>>>On Sat, 28 May 2022 18:01:30 +0200
>>>Rolf Mantel <news@hartig-mantel.de> wrote:
>>
>>>>Cloud servers enabled a REST-ful programming paradigm, without which 90%
>>>>of our internet-enabled mobile phone apps would not be feasible.
>>>
>>>Yes they would. Apps are just programs, they have little to do with the web
>>>(steve jobs made sure of that) unless they're written in javascript.
>>
>>How do you think mobile apps communicate with their data source, then?
>
>TCP. Next.

And what do you think runs over TCP when they use it?

Mark

On Mon, 30 May 2022 21:09:53 +0100
Mark Goodge <usenet@listmail.good-stuff.co.uk> wrote:
>On Mon, 30 May 2022 19:38:26 -0000 (UTC), muttley@dastardlyhq.com wrote:
>
>>On Mon, 30 May 2022 20:25:03 +0100
>>Mark Goodge <usenet@listmail.good-stuff.co.uk> wrote:
>>>On Sun, 29 May 2022 07:44:51 -0000 (UTC), Muttley@dastardlyhq.com wrote:
>>>
>>>>On Sat, 28 May 2022 18:01:30 +0200
>>>>Rolf Mantel <news@hartig-mantel.de> wrote:
>>>
>>>>>Cloud servers enabled a REST-ful programming paradigm, without which 90%
>>>>>of our internet-enabled mobile phone apps would not be feasible.
>>>>
>>>>Yes they would. Apps are just programs, they have little to do with the web
>>>>(steve jobs made sure of that) unless they're written in javascript.
>>>
>>>How do you think mobile apps communicate with their data source, then?
>>
>>TCP. Next.
>
>And what do you think runs over TCP when they use it?

It could be any protocol. You do realise ones other than HTTP exist I
presume?

On Tue, 31 May 2022 07:55:38 -0000 (UTC), muttley@dastardlyhq.com wrote:

>On Mon, 30 May 2022 21:09:53 +0100
>Mark Goodge <usenet@listmail.good-stuff.co.uk> wrote:
>>On Mon, 30 May 2022 19:38:26 -0000 (UTC), muttley@dastardlyhq.com wrote:
>>
>>>On Mon, 30 May 2022 20:25:03 +0100
>>>Mark Goodge <usenet@listmail.good-stuff.co.uk> wrote:
>>>>On Sun, 29 May 2022 07:44:51 -0000 (UTC), Muttley@dastardlyhq.com wrote:
>>>>
>>>>>On Sat, 28 May 2022 18:01:30 +0200
>>>>>Rolf Mantel <news@hartig-mantel.de> wrote:
>>>>
>>>>>>Cloud servers enabled a REST-ful programming paradigm, without which 90%
>>>>>>of our internet-enabled mobile phone apps would not be feasible.
>>>>>
>>>>>Yes they would. Apps are just programs, they have little to do with the web
>>>>>(steve jobs made sure of that) unless they're written in javascript.
>>>>
>>>>How do you think mobile apps communicate with their data source, then?
>>>
>>>TCP. Next.
>>
>>And what do you think runs over TCP when they use it?
>
>It could be any protocol. You do realise ones other than HTTP exist I
>presume?

Of course they do, and they're all part of the web :-)

But, actually, most apps communicate with the mothership via HTTP these
days. One of the advantages of HTTP is that it supports encrypted
authenticated requests, which is generally necessary for an app. The
only significant exception is for chat apps, which typically use XMPP or
WebSocket for full duplex communication (something that HTTP still isn't
particularly well suited to).

WebSocket is an interesting one in this context, because, as the name
suggests, it's very much a part of the web despite not being HTTP. It's
supported by all the main browsers as well as Apache, IIS and Nginx on
the server side, and allows genuine two-way communication between
browser and server in a way that HTTP can only mimic.

Mark

Roland Perry <roland@perry.co.uk> wrote:
> In message <t654b7$asp$1@dont-email.me>, at 09:58:00 on Thu, 19 May
> 2022, Sam Wilson <ukr@dummy.wislons.fastmail.co.uk> remarked:
>> Roland Perry <roland@perry.co.uk> wrote:
>>> In message <t62ct9$eq$1@dont-email.me>, at 09:05:45 on Wed, 18 May 2022,
>>> Sam Wilson <ukr@dummy.wislons.fastmail.co.uk> remarked:
>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>> In message <t5ba04$7g1$1@dont-email.me>, at 14:55:00 on Mon, 9 May 2022,
>>>>> Sam Wilson <ukr@dummy.wislons.fastmail.co.uk> remarked:
>>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>>> In message <t5aobq$qbf$1@dont-email.me>, at 09:54:02 on Mon, 9 May 2022,
>>>>>>> Sam Wilson <ukr@dummy.wislons.fastmail.co.uk> remarked:
>>>>>>>> Ken <ken@birchanger.com> wrote:
>>>>>>>>> On Sun, 8 May 2022 18:03:44 -0000 (UTC), Sam Wilson
>>>>>>>>> <ukr@dummy.wislons.fastmail.co.uk> wrote:
>>>>>>>>>
>>>>>>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>>>>>>> In message <t53foa$lbk$1@gioia.aioe.org>, at 15:44:10 on Fri, 6 May
>>>>>>>>>>> 2022, Muttley@dastardlyhq.com remarked:
>>>>>>>>>>>> On Fri, 6 May 2022 14:56:52 +0100
>>>>>>>>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>>>>>>>>> In message <p99a7h1hses2g7a0tiul6q3bs8nio1ahk6@4ax.com>, at
>>>>>>>>>>>>> 14:40:34 on
>>>>>>>>>>>>> Fri, 6 May 2022, Recliner <recliner.usenet@gmail.com> remarked:
>>>>>>>>>>>>>>> In what sense is the "cloud" not a line? It still consists of
>>>>>>>>>>>>>>> physically
>>>>>>>>>>>>>>> connected components despite the marketing name.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Might there not be satellite, microwave or WiFi links
>>>>>>>>>>>>>> along the way?
>>>>>>>>>>>>>
>>>>>>>>>>>>> Shush! I wasn't going to raise that. However, especially on bad
>>>>>>>>>>>>> connections, the packets might not even be flowing through all
>>>>>>>>>>>>> the same routers from one second to the next.
>>>>>>>>>>>>
>>>>>>>>>>>> "Cloud" is nothing more than marketdroid terminology for the
>>>>>>>>>>>> internet. The latter sounds techy and scary, the former warm
>>>>>>>>>>>> and fluffy and doesn't scare off the dimwit techno
>>>>>>>>>>>> illiterates in the boardroom.
>>>>>>>>>>>
>>>>>>>>>>> Once again, you fail to distinguish between the cloud-services/cloud
>>>>>>>>>>> storage, and the stuff which connects it all together.
>>>>>>>>>>
>>>>>>>>>> Which, I will repeat, was never referred to as a cloud until
>>>>>>>>>> cloud services came along.
>>>>>>>>>
>>>>>>>>> Hmm. When I was putting together X.25 systems a long, long time ago we
>>>>>>>>> always used a cloud symbol in documents and informally called the X.25
>>>>>>>>> network the cloud. I'd almost forgotten that.
>>>>>>>>
>>>>>>>> Yes, we always drew switched networks, or anything else where
>>>>>>>> the topology was unknown or irrelevant, as a cloud, but any
>>>>>>>> reference to the name was always incidental and informal. I’m
>>>>>>>> pretty sure you won’t find any formal documentation referring
>>>>>>>> to the network as a/the cloud except in passing.
>>>>>>>
>>>>>>> Could you bracket that assertion within some dates?
>>>>>>
>>>>>> Late 80s to around the turn of the millennium.
>>>>>>
>>>>>> FWIW I would still understand “the cloud” to mean the services
>>>>>> provided rather than the means of accessing them.
>>>>>
>>>>> Sure, but I'm talking about *a* cloud. As was pointed out earlier, "The
>>>>> Cloud" is the trading name of a public wifi network.
>>>>
>>>> So when people talk about storing their data in the cloud they’re
>>>> referring to that wifi network?
>>>
>>> No, that's just a clash of a concept and a tradename. No-one would talk
>>> about storing their data "in the O2" for example (unless they worked in
>>> the Dome, of course).
>>
>> But similarly they don’t store their data in the Internet,
>
> Whoever said they did? You've caught the Muttley-virus which can't
> distinguish between cloud[hosted] services, and the cloud connecting
> those services with their users.
>
>> but in services which are accessed via the Internet.
>>
>> Perhaps we are talking past each other because you regard networked
>> services as being part of the network (“The Network Is The Computer”)
>> whereas I don’t.
>
> The network providing the services is BellHead. I'm very much a NetHead
> where the services are provided at the end points by an essentially dumb
> [apart from its ability to route] network.

Excuse me for returning late to this conversation. My experience was
always that none of the NetHeads of my aquaintance (and I’m one of them
too) ever referred to the thing that we drew as a cloud as The Cloud - that
term was reserved for the services which were accessed via the network.
YMMV.

Sam

--
The entity formerly known as Sam.Wilson@ed.ac.uk
Spit the dummy to reply

Certes <none@nowhere.net> wrote:
> On 24/05/2022 15:07, Roland Perry wrote:
>> In message <t6hvv8$1o2i$1@gioia.aioe.org>, at 07:03:05 on Tue, 24 May
>> 2022, Muttley@dastardlyhq.com remarked:
>>> On Mon, 23 May 2022 18:08:27 +0100
>>> Roland Perry <roland@perry.co.uk> wrote:
>>>> In message <t6ga1i$mag$1@gioia.aioe.org>, at 15:42:42 on Mon, 23 May
>>>> 2022, Muttley@dastardlyhq.com remarked:
>>>>> On Mon, 23 May 2022 13:19:23 +0100
>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>> In message <t6fjc9$1lsv$1@gioia.aioe.org>, at 09:15:53 on Mon, 23 May
>>>>>> 2022, Muttley@dastardlyhq.com remarked:
>>>>>>> On Sun, 22 May 2022 16:32:52 +0100
>>>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>>>> In message <t65s80$hvg$1@dont-email.me>, at 16:45:52 on Thu, 19 May
>>>>>>>> 2022, Tweed <usenet.tweed@gmail.com> remarked:
>>>>>>
>>>>>>>>> The real difference between modern Cloud based services and
>>>>>>>>> remotely
>>>>>>>>> provided services at some location other than where you are in
>>>>>>>>> days gone
>>>>>>>>> by, is that the latter were usually in one well known data
>>>>>>>>> centre with
>>>>>>>>> possibly a standby location. Customers of modern Cloud services
>>>>>>>>> usually
>>>>>>>>> have no idea where the data centre is (you might know the legal
>>>>>>>>> jurisdiction if required) and your data and workload can be
>>>>>>>>> splattered
>>>>>>>>> across multiple locations and even dynamically move location.
>>>>>>>>
>>>>>>>> Indeed so. And I have tired of repeating this self-evident truth to
>>>>>>>> Muttley, because he's just not listening.
>>>>>>>
>>>>>>> So you and all the other users always knew the physical location
>>>>>>> of whatever
>>>>
>>>>>>> mapped drives you used in your corporate office did you?
>>>>>>
>>>>>> The receptionist probably didn't (nor would we expect them to).
>>>>>>
>>>>>> Most of the more permanent users probably had some idea, and if
>>>>>> necessary could ask. "It's the third server down, the second rack, in
>>>>>> our suite in Telehouse".
>>>>>
>>>>> So if they didn't know it would be cloud but because some did it
>>>>> wasn't.
>>>>> Got it.
>>>>
>>>> If they didn't know, it might be because it was the server in the back
>>>> room at the office, or the second server down in the third rack at
>>>> Telecity.
>>>>
>>>> None of these answers imply it was therefore distributed within a cloud.
>>>
>>> You do realise distributed just means different computers?
>>
>> It's necessary, but not sufficient.
>>
>> Here's a formal definition picked with a pin:
>>
>>    Distributed computing (or distributed processing) is the technique of
>>    linking together multiple computer servers over a network into a
>>    cluster, to share data and to coordinate processing power.
>>
>> And such clusters are often colloquially referred to as a cloud.
>
> “A distributed system is one in which the failure of a computer you
> didn’t even know existed can render your own computer unusable”
> - Leslie Lamport

‘Ethernets are not only ‘‘distributed packet switches’’ but ‘‘distributed
single points of failure.’’’ — Boggs, Mogul and Kent

Sam

--
The entity formerly known as Sam.Wilson@ed.ac.uk
Spit the dummy to reply

Sam Wilson <ukr@dummy.wislons.fastmail.co.uk> wrote:
> Certes <none@nowhere.net> wrote:
>> On 24/05/2022 15:07, Roland Perry wrote:
>>> In message <t6hvv8$1o2i$1@gioia.aioe.org>, at 07:03:05 on Tue, 24 May
>>> 2022, Muttley@dastardlyhq.com remarked:
>>>> On Mon, 23 May 2022 18:08:27 +0100
>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>> In message <t6ga1i$mag$1@gioia.aioe.org>, at 15:42:42 on Mon, 23 May
>>>>> 2022, Muttley@dastardlyhq.com remarked:
>>>>>> On Mon, 23 May 2022 13:19:23 +0100
>>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>>> In message <t6fjc9$1lsv$1@gioia.aioe.org>, at 09:15:53 on Mon, 23 May
>>>>>>> 2022, Muttley@dastardlyhq.com remarked:
>>>>>>>> On Sun, 22 May 2022 16:32:52 +0100
>>>>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>>>>> In message <t65s80$hvg$1@dont-email.me>, at 16:45:52 on Thu, 19 May
>>>>>>>>> 2022, Tweed <usenet.tweed@gmail.com> remarked:
>>>>>>>
>>>>>>>>>> The real difference between modern Cloud based services and
>>>>>>>>>> remotely
>>>>>>>>>> provided services at some location other than where you are in
>>>>>>>>>> days gone
>>>>>>>>>> by, is that the latter were usually in one well known data
>>>>>>>>>> centre with
>>>>>>>>>> possibly a standby location. Customers of modern Cloud services
>>>>>>>>>> usually
>>>>>>>>>> have no idea where the data centre is (you might know the legal
>>>>>>>>>> jurisdiction if required) and your data and workload can be
>>>>>>>>>> splattered
>>>>>>>>>> across multiple locations and even dynamically move location.
>>>>>>>>>
>>>>>>>>> Indeed so. And I have tired of repeating this self-evident truth to
>>>>>>>>> Muttley, because he's just not listening.
>>>>>>>>
>>>>>>>> So you and all the other users always knew the physical location
>>>>>>>> of whatever
>>>>>
>>>>>>>> mapped drives you used in your corporate office did you?
>>>>>>>
>>>>>>> The receptionist probably didn't (nor would we expect them to).
>>>>>>>
>>>>>>> Most of the more permanent users probably had some idea, and if
>>>>>>> necessary could ask. "It's the third server down, the second rack, in
>>>>>>> our suite in Telehouse".
>>>>>>
>>>>>> So if they didn't know it would be cloud but because some did it
>>>>>> wasn't.
>>>>>> Got it.
>>>>>
>>>>> If they didn't know, it might be because it was the server in the back
>>>>> room at the office, or the second server down in the third rack at
>>>>> Telecity.
>>>>>
>>>>> None of these answers imply it was therefore distributed within a cloud.
>>>>
>>>> You do realise distributed just means different computers?
>>>
>>> It's necessary, but not sufficient.
>>>
>>> Here's a formal definition picked with a pin:
>>>
>>>    Distributed computing (or distributed processing) is the technique of
>>>    linking together multiple computer servers over a network into a
>>>    cluster, to share data and to coordinate processing power.
>>>
>>> And such clusters are often colloquially referred to as a cloud.
>>
>> “A distributed system is one in which the failure of a computer you
>> didn’t even know existed can render your own computer unusable”
>> - Leslie Lamport
>
> ‘Ethernets are not only ‘‘distributed packet switches’’ but ‘‘distributed
> single points of failure.’’’ — Boggs, Mogul and Kent
>
> Sam
>

But surprisingly it all works very well most of the time.

Tweed <usenet.tweed@gmail.com> wrote:
> Sam Wilson <ukr@dummy.wislons.fastmail.co.uk> wrote:
>> Certes <none@nowhere.net> wrote:
>>> On 24/05/2022 15:07, Roland Perry wrote:
>>>> In message <t6hvv8$1o2i$1@gioia.aioe.org>, at 07:03:05 on Tue, 24 May
>>>> 2022, Muttley@dastardlyhq.com remarked:
>>>>> On Mon, 23 May 2022 18:08:27 +0100
>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>> In message <t6ga1i$mag$1@gioia.aioe.org>, at 15:42:42 on Mon, 23 May
>>>>>> 2022, Muttley@dastardlyhq.com remarked:
>>>>>>> On Mon, 23 May 2022 13:19:23 +0100
>>>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>>>> In message <t6fjc9$1lsv$1@gioia.aioe.org>, at 09:15:53 on Mon, 23 May
>>>>>>>> 2022, Muttley@dastardlyhq.com remarked:
>>>>>>>>> On Sun, 22 May 2022 16:32:52 +0100
>>>>>>>>> Roland Perry <roland@perry.co.uk> wrote:
>>>>>>>>>> In message <t65s80$hvg$1@dont-email.me>, at 16:45:52 on Thu, 19 May
>>>>>>>>>> 2022, Tweed <usenet.tweed@gmail.com> remarked:
>>>>>>>>
>>>>>>>>>>> The real difference between modern Cloud based services and
>>>>>>>>>>> remotely
>>>>>>>>>>> provided services at some location other than where you are in
>>>>>>>>>>> days gone
>>>>>>>>>>> by, is that the latter were usually in one well known data
>>>>>>>>>>> centre with
>>>>>>>>>>> possibly a standby location. Customers of modern Cloud services
>>>>>>>>>>> usually
>>>>>>>>>>> have no idea where the data centre is (you might know the legal
>>>>>>>>>>> jurisdiction if required) and your data and workload can be
>>>>>>>>>>> splattered
>>>>>>>>>>> across multiple locations and even dynamically move location.
>>>>>>>>>>
>>>>>>>>>> Indeed so. And I have tired of repeating this self-evident truth to
>>>>>>>>>> Muttley, because he's just not listening.
>>>>>>>>>
>>>>>>>>> So you and all the other users always knew the physical location
>>>>>>>>> of whatever
>>>>>>
>>>>>>>>> mapped drives you used in your corporate office did you?
>>>>>>>>
>>>>>>>> The receptionist probably didn't (nor would we expect them to).
>>>>>>>>
>>>>>>>> Most of the more permanent users probably had some idea, and if
>>>>>>>> necessary could ask. "It's the third server down, the second rack, in
>>>>>>>> our suite in Telehouse".
>>>>>>>
>>>>>>> So if they didn't know it would be cloud but because some did it
>>>>>>> wasn't.
>>>>>>> Got it.
>>>>>>
>>>>>> If they didn't know, it might be because it was the server in the back
>>>>>> room at the office, or the second server down in the third rack at
>>>>>> Telecity.
>>>>>>
>>>>>> None of these answers imply it was therefore distributed within a cloud.
>>>>>
>>>>> You do realise distributed just means different computers?
>>>>
>>>> It's necessary, but not sufficient.
>>>>
>>>> Here's a formal definition picked with a pin:
>>>>
>>>>    Distributed computing (or distributed processing) is the technique of
>>>>    linking together multiple computer servers over a network into a
>>>>    cluster, to share data and to coordinate processing power.
>>>>
>>>> And such clusters are often colloquially referred to as a cloud.
>>>
>>> “A distributed system is one in which the failure of a computer you
>>> didn’t even know existed can render your own computer unusable”
>>> - Leslie Lamport
>>
>> ‘Ethernets are not only ‘‘distributed packet switches’’ but ‘‘distributed
>> single points of failure.’’’ — Boggs, Mogul and Kent
>>
>> Sam
>>
>
> But surprisingly it all works very well most of the time.

To be fair they were writing in 1988 about shared medium coaxial ethernet.
10base-T wasn’t standardised until 1990. UTP and switches have
revolutionised things since then.

https://www.hpl.hp.com/techreports/Compaq-DEC/WRL-88-4.pdf

Sam

--
The entity formerly known as Sam.Wilson@ed.ac.uk
Spit the dummy to reply