There are two cases for changing SSL/TLS cipher settings
I've recently been reading a discussion about TLS cipher settings fora program. The suggestion was made that there should be some sort ofabstract high level setting for this, one that would be all that mostpeople should ever need (partly this was proposed on the entirelysensible grounds that TLS cipher settings are insanely difficult to getright unless you're a real expert). While this is perhaps a good ideain theory, I'm not sure it's going to work in practice. You see
Disused addresses and the impact of spam
Here's something that I've slowly realized, at least about myself: theannoyance and impact of a given volume of spam is disproportionatelylarge on otherwise idle and disused email addresses. The less realvolume the email address gets, the worse spam feels regardless ofvolume. I have some old email addresses that get perhaps one real emailevery six months and perhaps a spam email every week or two, and I findit teeth-grinding. Meanwhile I'd only find that (low) volume of spam
Why I'm not looking for any alternatives to iSCSI for us
Broad scale distributed storage systems such as Ceph are an in thingthese days (at least in some quarters). A while back a commentator on this entry suggested looking at themas an alternative to our use of iSCSI and I've been mulling over myreaction since then. Let me put it simply: my reaction is stronglynegative. The short reason why is that I see no compelling benefits andall alternatives appear to involve more complexity and magic.
Let's assume that the setuid issue can be dealt with
The importance of small UI tweaks (for me), dmenu edition
The basic behavior of dmenu is that itreads a bunch of things from standard input then puts up a windowwhich displays those things (or as many of them as fit) and lets youtype things with autocompletion from the list. The standard version ofdmenu puts as much as possible as it reads from standard input into thedisplay area and as a result if you have a lot of completion items youget a very cluttered display area. Because I don't like this sort ofclutter I spent a long time
Revisiting some bits of ZFS's ZIL with separate log devices
Back in this entry I described how the ZFS ZIL mayor may not put large writes into the ZIL itself depending on variousfactors (how big they are, whether you're using a separate ZIL device,and so on). It turns out that I missed one potentially important factor,in fact one that affects more than large writes.
If you're using a separate log device, ZFS will normally put all writedata into the ZIL (on the presumption that flushing data to the SLOGis faster than
How DWiki parses its wikitext (part 1)
I have a long-standing view that parsing wikitext is both hard and notdiscussed very much; if for some reason you have to write a wikitextparser, there are very few good examples to easily crib from. So Ifeel like describing at least some stuff about how DWiki parses itswikitext dialect .
(Most of current DWiki wikitext parsing is not my work; the core codewas generously improved and rewritten by Daniel Martin many years ago.)
The first big thing is that DWiki actually effectively has two separateparsers
Some pain points of parsing wikitext (and simplifications that avoid them)
I've talked before in general about things that makeparsing a general wikitext hard or at least inobvious but I've nevergiven specific examples. Today I want to do so and as an associatedissue talk about what simplifications and shortcuts you can take to makeit easier.
First, let's talk about errors. Specifically, that in parsing generalwikitext there aren't errors ; you've always got tokeep on producing output. In a wikitext dialect where * is theemphasis delimiter, consider the
Sun's NeWS was a mistake, as are all toolkit-in-server windowing systems
One of the great white hopes of the part of the Unixworld that never liked X Windows was Sun's NeWS . Never mind all of its practicalflaws, all sorts of people held NeWS up as the better way and the brightfuture that could have been if only things had been different (by whichthey mean if people had made the 'right' choice instead of settling forX Windows). One of the reasons people often give for liking NeWS is thatit put much of the windowing toolkit into the
An interesting bug with module garbage collection in Python
In response to my entry on what happens when modules are destroyed , @eevee shared the issue that started itall :
@thatcks the confusion arose whenthe dev did `module_maker().some_function_returning_a_global()` andgot None :)
In a subsequent exchange of tweets, we sorted out why this happens.What it boils down to is a module is not the same as the module'snamespace and functions only hold a reference to the modulenamespace, not
What happens when CPython deletes a module
The first thing to know about what happens when Python actually deletesand garbage-collects a module is that it doesn't happen very often; infact, you usually have to go out of your way to observe it. The two waysI know of to force a module to be deleted and garbage collected are toend your program (which causes all modules to eventually go throughgarbage collection) and to specifically delete all references to amodule, including in sys.modules . People don't do the latter