Not necessarily all evil. Only text-book implementations require that one allocate individual nodes. Most implementations I have seen in Linux and FreeBSD can pool allocate the nodes. One can also pull off neat tricks to append link chains in these implementations to expand the list without the need to reallocate memory as needed by arrays.

Pool allocation only addresses one of the problems with linked list; it still leaves you chasing pointers to iterate, it still decreases locality, and it still adds 4-16 bytes to every element in the container.

Meanwhile: you aren't really saying most people use pool-allocated lists, right? I see people hand-hacking malloc'd lists all the time. I rarely see custom pool allocators. It's what people were taught to do.

Here is a list of patents and industry vendors sued: http://intellectualventures.com/Libraries/Article_Reprints/F...

I looked at a couple of the security patents, where they're suing Symantec, Trend Micro, McAfee, and Checkpoint. These certainly seem to be acquisition rather than IV-developed ideas. So the methodology here (buy up undervalued patents and sue big companies) looks trollish to me.

People who can authoritatively talk about such attacks can't or don't talk. So, information out there is mostly anecdotal at best or just plain speculation. These cables add to that stack of information but from a much higher level of involvement. It may not be new or news but it's certainly from more credible source than what we usually hear from.

I'm objecting to the wording of the headline. I'm sorry if it seems like I'm arguing that there's no story at all.

The "China's official involvement in Google hack confirmed" headline is a serious overreach.

True. For the headline to be solid, these would have to be Chinese cables, not US ones. But this is probably closest we are going to get.

Another way is probably to use a Linux Kernel-like list: http://isis.poly.edu/kulesh/stuff/src/klist/

The kernel list approach also has the advantage that there's only one implementation of the list operation functions, whereas the one outlined in the article generates a new set of functions for each type.

Probably a dumb question: Are the reviews private (or only shared with friends) or public like other sites?

The reviews are public, but you choose a nickname to identify you (you're not identified by your Google account). The "friends" feature lets you get recommendations from those people. In addition, a friend sees your Google Profile name and picture, not just your review nickname.

Aikido

I would have liked to see him at FCC. Nevertheless a welcome change in appointments.

One of the big problems is there is no incentive for repeating or asserting previous results/findings. So even if someone is doubtful of an assertion made there is no incentive to follow up and verify the assertion in general. I don't think sharing code or secret data cleaning methods is going to bring much change unless someone is rewarded for repeating the results.

Reminded me of Epic Windows 98 Logon: http://i.imgur.com/JPxql.gif

Can we turn down this 'author-is-using-a-cheap-trick-to-attract-audience' volume around here. I don't think Bruce Eckel needs to attract audience so badly that he will do it purposefully. Granted, title is not reflective of the content but give the authors some benefit of the doubt. Let's do a quick Google search on authors before we write stuff like this.

You mean evaluate people based on who they are versus what they say? No thanks.

I interpreted the title to mean Google Groups was going away. I was not aware that there was a separate "data" concept.

That sounds like your lack of familarity with the software and not the author's showmanship.

If the title read: "Google Groups Shutting Down!" and they meant shutting down the pages and files part, then sure I'd see your argument. But literally all data in Google Groups will be destroyed. I actually thought the tittle meant his data was destroyed on his Google Group, much less all of them were being wiped. I think it is pretty big news.

> You mean evaluate people based on who they are versus what they say? No thanks.

I don't know. I usually find that knowing who people are often provides some context to what they are saying.

TFA has "data" in the title. You weren't aware of what it meant and jumped to a conclusion. I don't see how it follows that the author is being sensationalistic.