Live data from Hacker News

What's next Google? Dropping SMTP support?

eschnou.com

241–250 of 263 posts

Re: What's next Google? Dropping SMTP support?

#241
post #235

Earlier quoted context omitted.

Google Reader killed the RSS reader market by offering a solid product for free. You can't use the lack of other options as proof of anything except the depth of Google's pockets.

What does "killing the RSS reader market" (a point with which I agree, by the way) have to do with "turning against open standards"? If anything, your logic would argue that the harm to open standards happened when Reader was introduced , not when it was killed, no?

I'm not the original poster, so I can't say what he might mean by that.

But in my view, Reader's arc is a sign of their growing indifference to open standards and a love of walled gardens. Well, their walled garden.

My impression is that Reader was launched back in the days when they were just doing things for the fun of it. If a 20% time project does well, then you push it out. And then they cared about open standards, which is why you can OPML-export your reader subscriptions.

But even at the time they didn't think much about crushing the nascent RSS-reader market. The programmers wanted to build things, so they did. And they liked them to be open, so that's how they ended up.

In the last few years, though, they've shifted. Google Plus was a sign that the power shifted from the nerds to the suits. The suits see what Facebook has and want the same. In that period, they just neglected Reader to death. I doubt it was intentional, but they it suited their walled garden urges just fine.

Now they've killed it, because it doesn't even do that anymore. As many people have noted, RSS has lost a lot of ground to proprietary sharing. Will it make it back? Would it have lost it anyhow? Hard to say. But a big part of why it's hard to say is Google's dominance.

So trying to answer your last question as framed, I guess I'd say that the harm to open standards happened at launch and up until when they killed it. The announcement kicked off a bunch of innovation and interest in newsreaders, and I'm looking forward to seeing how that plays out.

Re: What's next Google? Dropping SMTP support?

#242
post #197

1. Google creates a service and gives it away for free. 2. Most alternative providers of the service die. The only ones that still provide the service must also give it away for free and leverage the service in some other way. 3. Google kills service. Someone might try to fill the void, but why would they? Google might just provide the service again for free. I wonder how much of the economy has been lost by giving a…

There's something wrong with the model where you can give something away for free so you can learn more about them to serve ads There's nothing wrong with the model while Google or another corp do it for their own money (unlike gov.). People are freely and willingly agree to see the ads in exchange to using free services. If you think they all wrong, that's just your personal opinion.

> If you think [X], that's just your personal opinion.

You're saying "I don't trust your opinion" without any justification beyond asserting that personal opinions have low status. If you suspect Markus can't back his claims up, then ask him to. Also remember that if you hold a contrary opinion, then one of you two (possibly both) is necessarily wrong as a simple matter of fact.

When Markus said "there's something wrong with X", I guess he's actually saying "X does harm". In the case of Gmail, I'm inclined to believe that it does harm: better brain hacking (ads) through breaches of privacy —Google is effectively reading your mail. Also the "destroying the economy" effect may be quite real, though I think it's probably limited to short term damages.

> People are freely and willingly agree to […].

You're assuming people are rational. They're not. People freely and willingly agree to bamboozle themselves all the time, sometimes creating a network effect that will make it easier for more people to bamboozle themselves. Some people gamble too much and owe money to a Mafioso. Others join the Scientology. Or start doing drugs in a moment of weakness. Or use proprietary software, and eventually lose data. Etc, etc.

About the specific case of "cloud" centralized services (web mail, video streaming, blogs, social networks…), the main harm is a progressive loss of privacy, and increased dependence. I think it all stemmed from the thought that people are consumers, not creators. Technically this got translated by "people hardly need to upload". Hence the 'A' in "ADSL". Symmetry was sacrificed at the altar of download rate. Since then, any hope to have a decent server at home was basically crushed. Some providers even prevent you to host a server, or to send e-mail (they typically force you to go through their crappy, rate-limited relays).

Anyway, people did create. They wrote. They made videos. They shared their life with friends. And they continued to send e-mail. Well, the only practical solution to do all that was to use convenient centralized services. No way any given person would have enough bandwidth to host one's popular video. So it all goes to YouTube, creating a serious imbalance in the data transfer rates that makes some network operators uneasy.

Oh, there's an exception: Bit-torrent. It is an incredibly convenient and scalable way to share big files on the internet. And it's quite popular. Alas, because of asymmetric bandwidth, it doesn't work nearly as well as it should. Which is basically why we used MegaUpload —until "Big IP" shut it down. With symmetric bandwidth, it would have had no use, and Big IP couldn't have annoyed so many people.

We could have had symmetric bandwidth, convenient servers at home and all the privacy and independence that they can give us. However the market, people's preference for short term convenience, and network effects conspired against it. And the worse is, it works even if each individual choice is rational. Take the web mails: it could be rational for you to use it, but then, you participate in a network effect that get people to only accept email from the big players. This has already started: mails sent from home are often dismissed as spam right away. Most such mail is spam, but the (possibly rational) rejection is hurting the legitimate users.

Re: What's next Google? Dropping SMTP support?

#243
post #226

Earlier quoted context omitted.

Given that Google was, essentially, the only meaningful provider of either a public federated XMPP service or a RSS feed reader[1], that seems like a pretty weak argument. If there's no community, was it ever really open? [1] Calling Google's 100% proprietary (albeit publicly documented) reader protocol an "open standard" is a bit of a stretch, FWIW. Google certainly isn't refusing to support RSS on its blog offering…

There are lots of RSS desktop clients. It's unfortunate that web apps are appearing more vulnerable to Google's monopolistic practices than desktop apps were to Microsoft's.

Microsoft's products cost money (minus piracy), and fought Free Software, which is also Gratis most of the time.

Google is already Gratis (the ads don't count as a price in most people's minds).

Re: What's next Google? Dropping SMTP support?

#244
post #214

Earlier quoted context omitted.

I think he's saying we should instead have text-only e-mail, because that's both standard and has a tiny attack surface.

That doesn't fix anything either. Many clients make URLs clickable (even with text-only emails) and even if those clients didn't do that, someone will just copy and paste a URL into the address bar without looking. Given that the average user isn't an expert on domain names, it's not actually that hard to trick them into thinking that face.book.com isn't the same as facebook.com (and even those that are savvy enough…

It's impossible to prevent all cases of social engineering with a purely technical solution, this is true. But that doesn't mean that we should disregard how much easier phishing attacks are with the advent of HTML mail. The goal here is not so much to stop social engineering dead in its tracks as to prevent some of the low-effort attacks that do happen.

Re: What's next Google? Dropping SMTP support?

#245
post #219

Earlier quoted context omitted.

| So SMTP and POP3/IMAP? I think that the 'clients to talk to servers' part is referring to sending email, unless you think the webmail is the only way to send email. POP/IMAP are for reading email from the receiving server, not for sending out new emails. | So NNTP? I was under the impression that most Usenet networks used UUCP.

> POP/IMAP are for reading email from the receiving server, not for sending out new emails. They probably shouldn't be, though, there's an IMAP extension for submission. > I was under the impression that most Usenet networks used UUCP. No, NNTP. UUCP hasn't been used seriously for over a decade. Yours, news.ycombinator.com!ethomson

  | No, NNTP. UUCP hasn't been used seriously for
  | over a decade.
Looked it up and it seems you are mostly[1] right. Some time ago, I picked up the notion that most large Usenet networks synced articles between each other using UUCP, but clients used NNTP to pull down articles for reading. Not being a very avid Usenet user, I just took this at face value (also not knowing much about UUCP, like the fact that it used Layer 1 or Layer 2 protocol prior to TCP/IP).

[1] According to Wikipedia, UUCP was still in use in 2006 for at least one software product.

Re: What's next Google? Dropping SMTP support?

#246
post #223

Earlier quoted context omitted.

You can't put the genie back in the bottle, unfortunately. Most people use clients that support HTML right now, and a new format that doesn't support some sort of rich text decoration (like HTML) isn't going to become popular enough to matter.

A ridiculously cheap and powerful alternative to HTML (or any other static format for that matter), would be a bytecode format intended to be executed in a sandboxed virtual machine. Possibly something like Flash. That should give you all the bells and whistles you could possibly want with a document format. Now, there still are some loose ends about various screen sizes, preferred font family and size, and of course…

Really? Flash is more secure than HTML? Maybe we should just embed Java applets in emails? This still doesn't solve the "click a link that takes them elsewhere" issue that was brought up.

Re: What's next Google? Dropping SMTP support?

#247
post #156

Earlier quoted context omitted.

Considering that they only enabled federation again a little while ago it's not like that's something new.

They enabled federation months after Google Talk came out actually.

And then they disabled it a good while ago except for a few servers that they had whitelisted, thus my "again".

Re: What's next Google? Dropping SMTP support?

#248
post #245

Earlier quoted context omitted.

> POP/IMAP are for reading email from the receiving server, not for sending out new emails. They probably shouldn't be, though, there's an IMAP extension for submission. > I was under the impression that most Usenet networks used UUCP. No, NNTP. UUCP hasn't been used seriously for over a decade. Yours, news.ycombinator.com!ethomson

| No, NNTP. UUCP hasn't been used seriously for | over a decade. Looked it up and it seems you are mostly[1] right. Some time ago, I picked up the notion that most large Usenet networks synced articles between each other using UUCP, but clients used NNTP to pull down articles for reading. Not being a very avid Usenet user, I just took this at face value (also not knowing much about UUCP, like the fact that it used La…

Some people apparently use UUCP-over-SSH to get mails from their remote mailservers to their local ones, because, you know, fetchmail is evil or something.

Re: What's next Google? Dropping SMTP support?

#249

Earlier quoted context omitted.

There are lots of RSS desktop clients. It's unfortunate that web apps are appearing more vulnerable to Google's monopolistic practices than desktop apps were to Microsoft's.

Microsoft's products cost money (minus piracy), and fought Free Software, which is also Gratis most of the time. Google is already Gratis (the ads don't count as a price in most people's minds).

I was referring more to IE, which was bundled with the OS to circumvent Netscape's business model.

Re: What's next Google? Dropping SMTP support?

#250
post #3

Don't forget, they tried "open" with Google Wave, which was a total flop. Developers did not buy into it. You can't expect to have it both ways, especially since Apple has proven that a closed infrastructure (walled garden) works.

They've tried "closed" with Google+, which is also a total flop.

Yes, but they haven't realized that yet!
Post reply on HN