Computing desk
< June 30	<< Jun \| July \| Aug >>	Current desk >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

July 1

Open pdf in acrobat not in browser

When I click on a pdf link on a website, I want it to open in acrobat (reader DC) not in the browser. I have googled (and wikipedia RD searched) and all I can find is an instruction to associate pdf with acrobat. However pdf IS associated with acrobat, and if I click on a pdf in windows explorer, it does open in acrobat. To confirm this, I changed it to edge and then back to acrobat - no improvement. It only gets it wrong when launched from a browser link. The problem occurs in both Edge and Firefox (I haven't tried chrome yet). I have a laptop with Win 10. Any help gratefully received, thanks. -- SGBailey (talk) 06:37, 1 July 2020 (UTC)[reply]

For Firefox this can be done by going to options and then the applications section. Here you can change the default (firefox) to download or to windows default.

For Chrome I found:

1. Open Chrome and type "about:plugins" into the omnibox at the top.

2. Find Chrome PDF Viewer in the list and click the Disable link to prevent PDFs from loading within Chrome.

for Edge:

1. Click on the 3 dots at the top right of Edge. (((And then select settings edit by SGB)))

2. Click on Site Permissions on the left.

3. Click on PDF documents.

4. Toggle on Always open PDF files externally.

For reference, I used the search terms "disable pdf viewer" with edge/chrome/firefox behind it. Rmvandijk (talk) 08:00, 1 July 2020 (UTC)[reply]

That worked for Edge. Thanks. Yet to try firefox & chrome. BTW, in edge it doesn't "open", it comes up with a tiny tab at the bottom left that you can click on to get it to "open" - any idea if you can make the original click "open" the document - I suspect it won't do it. -- SGBailey (talk) 09:01, 1 July 2020 (UTC)[reply]

Hmm, I don't generally use edge and I believe it is now moving to a chromium based browser. I still have the old one so I can't check. Rmvandijk (talk) 11:57, 1 July 2020 (UTC)[reply]

IP ranges

mw:Help:Range blocks has an element that I don't understand: The first and last numbers of any block are reserved for network communication. So a 30-level block, with four addresses, has only two that aren't reserved. Take this group of addresses:

148.20.57.0
148.20.57.1
148.20.57.2
148.20.57.3
148.20.57.4
148.20.57.5
148.20.57.6
148.20.57.7

We could do two 30-level blocks (0-3 and 4-7), and only four of the addresses (1, 2, 5, and 6) would be usable. But if we do a 29-level block, encompassing all eight addresses, six of the eight will be usable, and only 0 and 7 would be unusable. Why does the availability of 3 and 4 depend on the length of the range? Why does the length of a range influence whether an IP address is reserved for network communication (maybe ICANN?) or whether it's available for normal use? Nyttend backup (talk) 14:59, 1 July 2020 (UTC)[reply]

Nyttend backup, I assume by 'level' you mean bits, as in 148.20.57.0/30. The statement in the range blocks help is not necessarily true. There is a lot of network equipment these days that do not need broadcast or network addresses, and therefore can treat some or all of those numbers as usable for host addressing.

The reason an all-zeroes address is historically unusable is because it is the network address itself. It is used to denote the network rather than a host on it. Network administrators would use it for such purposes.

An all-ones address is typically unusable because it is considered a broadcast address. If you ping the all-ones address then ping would hopefully use the broadcast Ethernet address to reach all hosts on the LAN. This is not observed so much anymore, although the broadcast may still be reserved in some cases.

As you can see, because IP addresses depend on bits set or unset in a mask, the all-ones or all-zeroes addresses can change depending on the length of the mask. In 148.20.57.0/30, that is a 2-bit host mask and so if 2 bits are both 0, or 2 bits are both 1, then that produces the "unusable" addresses, and so forth. Elizium23 (talk) 19:51, 1 July 2020 (UTC)[reply]

Sorry, by X-level I meant an /X range, e.g. "we could do two /30 blocks..." 148.20.57.0 is 10010100.00010100.00111001.00000000. So since you're talking about all-zeroes or all-ones, does that apply only to 255.255.255.255 and 0.0.0.0? Sorry for my confusion; I've just read the rangeblocks stuff time and time again, and never understood it much. Nyttend backup (talk) 11:23, 2 July 2020 (UTC)[reply]

The range block will block all the numbers in the subnet specified. The all zeros and all ones are within the subnet, so 10010100.00010100.00111001.00000000 is not usable for a network device, and 10010100.00010100.00111001.00000011 is not usable in that /30 subnet, as they are the first and the last in the range. You can turn that /30 into a bit mask that you "and" with the address 00000000.00000000.00000000.00000011 to get the number within the subnet, from 00 to 11. Graeme Bartlett (talk) 12:26, 2 July 2020 (UTC)[reply]

Nyttend backup, when I say "all zeroes" or "all ones" I am referring to the host portion of the address.

In 148.20.57.0/29, 3 bits are in the host portion, so the network address is 148.20.57.0 and the broadcast is 148.20.57.7.

In 148.20.57.0/30, 2 bits are in the host portion, so the network address is 148.20.57.0 and the broadcast is 148.20.57.3. Elizium23 (talk) 20:47, 4 July 2020 (UTC)[reply]

Also I am not sure why you are asking about usable addresses in the context of blocking. If you block a range, the whole range is blocked, usable and unusable alike. Elizium23 (talk) 20:50, 4 July 2020 (UTC)[reply]