Re: Call for Community Feedback: Retiring IETF FTP Service

[Julian Reschke <[e-mail]>]

Am 01.12.2020 um 13:33 schrieb tom petch:
> On 30/11/2020 17:41, Julian Reschke wrote:
>> Am 30.11.2020 um 18:27 schrieb tom petch:
>>> ...
>>> All the CSS preceding the text; well, that is what I got this morning
>>> when prompted by Julian's post but tomorrow, well, tomorrow is another
>>> day!
>>> ...
>>
>> On what URL?
>
> Julian
>
> Apologies for a tardy response - I am on European time and so much of
> the discussion is when I am in bed.

So am I :-)

> A bigger apology.  I used the URL you gave of
> https://www.rfc-editor.org/rfc/rfc7230.txt
> except that, when I look at the History of my web usage yesterday, I
> omitted the final .txt.  Repeating it with the .txt I get just the text
> and no preceding CSS.  So that was a red herring, omit the .txt and
> 'Save as' .txt saves all the CSS first; odd, but I know not to do that
> again. (The logic escapes me but then this is HTTP/HTML whose logic
> often does:-)

<https://www.rfc-editor.org/rfc/rfc7230> is an HTML resource (which just
happens to use a fixed space font). So you're asking IE to save an HTML
page as text file, and that's what it does in that case.

> The loss of formatting when using HTTP/HTML is still an issue for me; I
> recall it being an issue many years ago but not one I see with ftp://
> I access
> https://www.ietf.org/ietf-ftp/rfc/rfc5347.txt
> (a  URL most of which I took from an e-mail from Roman) and
> Save target as
> get a message that the download was successful (with .pdf I usually get

Plain text published by the RFC Editor does not have CRs in it (just
LFs), so you need to use a text editor that groks this format (Notepad
does this nowadays, and Wordpad has forever).

> an error message at this point) am invited to open it, do so and have
> lost the formatting, getting just a stream of characters.  I lack a hex

I just did a "save as" in IE with
<https://www.rfc-editor.org/rfc/pdfrfc/rfc7230.txt.pdf> and did not get
an error.

> editor at present (I am seeking to fix that) so do not know what is
> there but suspect it is the long standing issue of terminating lines
> with <CRLF> or something else.
>
> I am sure that given time I will solve most of my problems, except,
> probably, the loss of the original date for an RFC (that date is part of
> my modus operandi).
>
> The ongoing issue is, as others have commented on, I never know when and
> what I am going to encounter next with HTTP/HTML, as with my inability
> to signout the other day, or the Local Authority PC that insists that I
> have to stop what I am working on and look at the latest e-mail to
> arrive.  I see HTTP/HTML as not as stable as I want it to be as the
> basis for work, something I think I see at Enterprises where I work,
> where HTTP/HTML is always used, but for pushing out information and
> getting feedback, not for operational systems.
>
> So thank your for taking an interest.
> ...

Of all the issues you reported, the only thing I can reproduce is that
when saving text from the browser, the last modified time is indeed lost.

"curl" will allow you to keep it, for instance with

   curl --remote-time https://www.rfc-editor.org/rfc/rfc7230.txt -o foo.txt

So choose the right tool for the task, and there shouldn't be problem.

Best regards, Julian