The free software answer to the microsoft nightmare becoming OUR nightmare

Posted 26 Sep 2005 at 04:21 UTC by lkcl Share This

A recent slashdot and news.com article highlights the ten year nightmare for microsoft: the Internet taking over in a cheaper way, in which microsoft desktops cannot compete. My question to the free software community is: what is the answer (not to microsoft, as everyone keeps trying to fight microsoft on THEIR terms) but what is the answer to microsoft's nightmare also becoming OUR nightmare?

original news.com report

My concern: that, just as microsoft is left behind by google and the other upstarts, so will the free software desktop.

Just to clarify, so as to save people time on this issue:

1) i am fully aware that free software comprises a large amount of the SERVER infrastructure of google. this is not what i am asking about: i am asking about what SERVICES the free software community can provide that will result in free software as a SERVICE not falling behind what google etc. offer.

2) free software desktop software is free: for the sake of this article let us either assume that free software has "won", or that the cost of PCs are discounted with sufficient windows software and that the discount is sufficient incentive to NOT offset against the "inconvenience of unfamiliarity" in order to justify switching to free software, assume viruses don't exist, blah blah - assume that all that is irrelevant, and that "windows" is in the same boat as "free software" when it comes to online "services".

as developers, our computing and internet requirements are in stark contrast to everyone else (to the point where if you tell someone you're "online" for ten hours a day instead of "at a desk" for ten hours a day, they believe you need some serious help)

we all know and love wikis, mailing lists, irc, developer forums and possibly even, if we're old enough, newsgroups.

such resources and the forums we visit with those resources are completely useless to "ordinary" people, who want to be able to blog to their heart's content, watch telly, meet new people or yap to their friends, keep up with the latest real-world and fantasy-world and online-world fashions... you name it, it don't come _close_ to what we use the internet for, day-to-day, day-in, day-out.

the bottom line: where is the free software community's "MSN" and where is the free software community's "google bar"????

we are good at developing "online" tools for our OWN purposes, to improve our ability to do our own jobs: what "internet-interacive" tools should the free software community work on in order to enable - or better yet "empower" (an awful but unfortunately appropriate amehican word) - the rest of the world online, using free software at both the desktop and the server end?


no freedom (to modify) in web services, posted 26 Sep 2005 at 11:06 UTC by lkcl » (Master)

i received this reply by email and was kindly given permission to post it on, here. no explicit permission was given to mention the poster's contact details so i've left them out.

I think the main problem is that there is no "free software" in services. There is not the same freedom with a web service as there is on your own personal computer. "Services" is more than just a google bar.

Services means infrastructure. Someone has to run the services, pay for the bandwidth, do backups, be hooked up to a pager if the system goes down, etc.

And no administrator is going to let the random user, even if he's a geek, to edit the live source code of his website. Perhaps accept patches, but patches are applied to software, and software is not services.

It's like patching your own version of a php bulletin board, and testing on your system. Anyone can do that now, but it's not a service. The service is the community, the brand, the website, the business. It may make use of that php software, but it's a lot more.

So I'm not sure the whole nightmare is an issue. And if it is an issue, I don't think we're asking the right question.

Just my $0.02.

- Chris

distributed services, posted 26 Sep 2005 at 12:07 UTC by lkcl » (Master)

chris, hi, thanks for responding. your reply needs careful thought (at least by a donut like me :)

my thoughts - not in any particular order:

1) no, i don't believe the free software community is asking the right questions, and yes, of course, due to the massively distributed and mostly part-time volunteer nature of what we are and what we do, we're not a lean mean profit-making machine. therefore, microsoft's nightmare of being "beaten" by the google-bar syndrome and the whole concept of web services isn't such a big deal. maybe it's novell, ibm, hp, redhat etc.'s big deal, but it's not _our_ big deal.

what i _am_ concerned about is the free software community focussing on the wrong things and side-lining themselves into irrelevancy. missing opportunities by focussing on the glory of the chase for several years - going for the technical enjoyment of a challenge, without actually thinking first "hey, yeah, hundreds of thousands of ordinary people could really do with this stuff, let's DO IT, and FAST!".

at _least_ microsoft thinks ahead and goes "what technologies can we create that will, in N months or N years time, make a big difference to people's lives?". some of those things they may even get it right. i shouldn't attempt to point out _what_ they get right because it will be a matter of opinion, outside the scope of this article, and i will only get myself into deep doodoo :)

2) regarding your point that services require infrastructure: with carefully designed distributed systems, no they don't!!

imagine this: a self-backing-up distributed web service, that encrypts your data and automatically blops it out onto oh i dunno say freenet (whilst at the same time also adding your computer to the freenet infrastructure).

another example: let's say you wanted to do something mad like create a distributed search engine (as a free software competitor to google): there's nothing wrong with thinking ahead and using a peer-to-peer filesystem as the storage mechanism for the indexing information, and when someone does a search, do a lookup on the peer-to-peer filesystem, find out who's cached the page (which would be stored, again, in a file you download from the peer-to-peer filesystem) and then do an HTTP referral to the web service running on some other person's remote computer in order to serve the information from the cache.

i'm babbling, and spouting keywords here in a string of brainstorming nonsense but you get the idea: you don't _have_ to have centralised web services, you don't _have_ to have single points of failure that require 100% uptime and someone 100% to monitor them: even if you did, it would be possible to build a distributed monitoring system to notify a stack of registered volunteers across different timezones, telling them to scramble and why.

services like skype and bittorrent demonstrate the success of distributing information, processing power and communication quite nicely.

TOR - tor.eff.org also provides a good example by distributing network packets, but even that, in order to really take off, needs to be "bundled" with something that makes decent use of it (like skype quietly makes use of its subversive peer-to-peer communications infrastructure to ferry contact details, files, instant messaging and voice calls).

oh - and to be _really_ effective, TOR also needs to automatically self-configure (firewall busting) like skype does by routing incoming packets down outgoing connections initiated from inside your firewall (NAT sucks, so blow it).

TOR, BT, skype - these are all peer-to-peer services (god i hate the phrase peer-to-peer it's _so_ tacky *sigh*).

3) the only areas where the FOSS community takes a lead is in "subversive" and "contraversial" areas such as freenet, bittorrent and TOR, that tend to make ordinary people nervous: "here's some cool technology that makes an ethical or freedom-of-speech point but isn't actually useful to you for your day-to-day computing needs, and it you could get into trouble if you played with it".

yes these are technically services, yes they help solve some people's needs, but they are, i believe, simply tools in the chain: they're not _actually_ doing lots and lots of people an actual "service".

in fact, our own technical competence is doing them a DIS-service!

4) i _love_ your "so what _is_ the right question to ask" point.

* what do "real" people _need_ that we're not providing them?

* if we _are_ providing ordinary people with the best we can offer, then how can it be done better?

* the controversial and subversive areas aside (such as TOR), why do we keep following in other people's footsteps - gnutella takes on kazaa; openoffice takes on office; mozilla takes on IE; mono takes on .NET; samba takes on MS filesharing.

instead, we should be changing the technological landscape in a far-reaching direction that causes the big companies to follow _us_ for a change. as a community, we have the technical skills to pull off the biggest heist of this planet's known history, let's get off our fat backsides, work out what it is and DO it!!!

taking a stand, posted 26 Sep 2005 at 12:29 UTC by lkcl » (Master)

another reply from a friend of mine: whaa! i thought he was talking about _me_ being work-shy and ill-disciplined for a minute. wow. some harsh criticism, but it's from a successful businessman who is bewildered by the free software community's self-evident lack of direction / results / business-orientated goals and achievements. i believe that he sees only that the FOSS community glorifies and worships technical leadership over-and-above business sense (and for the most part, i personally also believe they do so over-and-above social responsibilities, too ... but then i'm one of the "Lost" too so what do i know :)

Of course it may just be the cry of the work-shy and ill-disciplined who are afraid of taking any responsibility for their own actions ... the point is that the "free community" has a chance of taking a leadership role in some fields and must change from being a protest organsiation to a delivery one ...

This may mean that 80% of the Free players would be lost (or better they form .. after Unix and Linux ... Nix2offer) because their presence was entirely to do with negative reasoning and they were never suited to any real responsibility !!!

Please stand up those who are the real backbone and cut away the hangers-on in order that a real contribution can be made and, just as important, sustained ...

spotted a diary entry, posted 26 Sep 2005 at 19:44 UTC by lkcl » (Master)

joolean, hi, i hope you don't mind me cross-referencing your relevant diary entry here.

the points you raise are very relevant - because you highlight several points which indicate flaws in what i was trying to convey.

lkcl seems to be making the argument that services are the new desktop PC / operating system, which is not really a valid comparison.

are services the new desktop PC? i don't know - should they be? (raising issues and others like them is the point of the article, far be it for me to tell people what should and shouldn't be).

that a comparison of desktop PCs and services is not valid is, i believe, irrelevant when put into perspective: a more important question is: what CAN deliver what ordinary people really want and need from computer technology (if anything). can desktop PCs deliver? can services? can mobile phones? can toasters deliver? can sun java?

as a side-note, bear in mind that we are working with a highly disillusioned, disempowered and software-technologically backwards and inhibited world population, whose response to virus and spyware attacks is tending towards the "throw the computer away and buy a new one" response, people who, for the past FIFTEEN YEARS, have been provided with no other real alternative, as far as they are concerned.

it's interesting to note that you believe google to be open because people can write according to their APIs and the advertising will remain embedded: i believe this to be flawed. what's to stop me writing a google ad-blocker?

client-server architecture vs peer-to-peer., posted 26 Sep 2005 at 23:11 UTC by lkcl » (Master)

following up on your second diary post, joolean, hi, yes it's a bit of a pain the advogato posting thing: about a year ago advogato.org ran out of disk space and trashed one half of the two-way Certification cross-links from other peoples profiles to my profile: i am therefore at rather an odd loose end Cert-wise - sorry! anyway, onwards...

for one thing, the services you're talking about require management (additional content, maintenance of remote hardware, etc.)

ah? as a peer-to-peer service, remote hardware is required? :) yes - everybody's computers become part of the service _to_ everyone else. no, there isn't any maintenance cost, and you must ensure proper design such that there is no dependence on any one piece of remote, and unreliable, hardware - for ANYTHING.

i hate the word, i hate the phrase, but development of what is referred to as "peer-to-peer" services do not incur the same maintenance costs as traditional "web services": there is no single point of failure, there is no reliance on internet access to a single computing resource. they do, however, incur a risk - a risk that you didn't do your design work properly such that the whole project falls in a creaking heap at above 500 simultaneous users.

odd, posted 26 Sep 2005 at 23:16 UTC by lkcl » (Master)

an odd observation: two readers have been kind enough to read and comment on this evolving article, such that i am able to relate this: that i fail to see why "web services" must absolutely must be "centralised". surely there's enough peer-to-peer infrastructure now kicking about in order to provide useful, relevant and importantly _distributed_ web services.

i realise that this implies dropping an entire web server plus supporting infrastructure onto every machine that is involved in the peer-to-peer network. hey, but let's be realistic: in order to be successful, we're talking a windows program here: you _really_ need all of that 60gb drive??

clarification of definition of "service", posted 27 Sep 2005 at 20:50 UTC by lkcl » (Master)

in my initial post, i don't mean "service" as in the TECHNICAL sense.

i mean "service" as in "serving the public good", "service industry" etc.

I'm saying no to ad-investation, posted 28 Sep 2005 at 18:25 UTC by prozac » (Journeyer)

what's to stop me writing a google ad-blocker?

I do block some Google ads. Not that I dislike Google Syndication, but I dislike the HUNDREDS of Webpages I see that have ZERO content but instead are filled with KEYWORDS and LINKS just to get HITS -- and all of them now have a link to http://pagead2.googlesyndication.com/pagead/show_ads.js.

But to get back to this:

where is the free software community's "MSN"

I am certainly in the minority, but I shy away from any "community" site designed around an "advertising" model. The data on MSN and Yahoo are 50% advertising and 50% content (more or less, it would be interesting to actually count the bytes). An "FSN" (Free Software Network) that looked and acted like MSN would be a terrible thing in my opinion -- unless it ONLY had "advertising" for other Free Software companies, products and services.

I blocked all OSDN ads when I went to one of ther sites and saw a large prominent (read "in the way") ad for an SUV!

Call me unrealistic, stupid, whatever, but Free Software is supposed to be something "higher minded" than selling advertising. Does Source Forge exist to make money from advertising or does Source Forge sell advertising to exist?

I don't like Source Forge becuase the underlying design purposefully (as I experience it) makes people click an excessive amount of times -- to maximize page loading -- to mazimize advertisment viewing - just to navigate the aprticulars of a project.

I think that that is unethical.

Imagine going to a Users Group to find corporate sales people hanging out. Imagine the Users Group managers actively pursuing corporate sales people to come hang out?

*sigh*

where is the MSN, posted 29 Sep 2005 at 16:22 UTC by lkcl » (Master)

hi,

okay: so, assuming that your comments are a representational view, then whatever is created - by the free software community to "serve" ordinary people - it shouldn't have any advertising in it :)

well, assuming that a revenue stream is required (at all, which a peer-to-peer distributed system would be extremely unlikely to need - no server infrastructure but an initial DNS record) then that would leave data mining and user habits to farm blocks of information and sell..

hmmm --- i like the sound of FSN, hmmm...

regarding your point that FSN should have "only" free software advertising: that makes it of little use / little value to ordinary people.

advertising needs to be targetted: some people actually appreciate it and consider it to be a service!

i love slashdot - you can strip away everything and just get content: your choice.

The Real Danger, posted 1 Oct 2005 at 09:48 UTC by ncm » (Master)

The real danger to Free Software from the web-service computing world is twofold.

First, companies can use any amount of Free Software to compete against us, without giving back. This is both because of Apache license terms (RMS points out that most of the Apache "market share" noted by Netcraft are actually running proprietary forks) and because the present GPL only encumbers distribution, but not, in legal terms, "public performance". The GPLv3 might mitigate this danger, if it's not too weenie, and if it is taken up aggressively.

Second, and more worrisomely, if the generic browser becomes the target platform for applications, then the standard hardware target we need to keep our costs down may disappear. Even if (say) 10% of the gadgets that replace computers were to remain actually "open" and documented, there's no way the Free Software community could maintain the thousands of ports needed to boot and run on all of them. Debian manages fewer than a dozen ports, NetBSD not so many more. How many variations are there just on the ARM, MIPS, PPC, and SH instruction sets, as implemented in various cores, each needing its own compiler port? A company needs to keep up only one compiler (or a few), with no requirement to distribute it, but we need to keep them all up. A foretaste is the slew of sites like ipodlinux.org and wrt54g.org. How many projects like that can we maintain?

I don't want to sound alarmist, but I'm worried.

The Google Legacy, posted 1 Oct 2005 at 12:55 UTC by lkcl » (Master)

THe Google Legacy

this is the sort of thing that i am talking about: what is the free software community's answer to _this_ - massively distributed services.

according to this report, it appears that google has implemented 70% of an online interoperable version of microsoft office, and that google has implemented, in google maps, something akin to microsoft powerpoint.

i don't see an online distributed free software version of openoffice where i can save my files, go half way across the world, lose my laptop, go into an internet cafe and CARRY ON WORKING because actually the laptop is only used to log in to a distributed free software "office" network where my laptop just happened to be ONE entity in the node.

_that's_ what i'm talking about.

and no, it doesn't have to be browser-based, but browser-based should just be one option, for convenience.

if google is really going ahead with plans like this, microsoft is _dead_. and, to some extent, the free software projects like openoffice, kde and gnome will look pretty stupid and irrelevant, especially if you have to have heavyweight PC hardware to run them on and "thin clients", connected virtually exclusively to google networks, become the "norm".

FSN - The Free Software Network, posted 2 Oct 2005 at 21:19 UTC by prozac » (Journeyer)

I'd love to see an FSN.

However, take a look, technically, the two sites that lkcl referred to: news.com.com and informationweek.com. Those two sites have 80 to 90% non-content. The latter has a animated "pop-down" advertisement that blocks their own content. Those site hare -- to me -- perfect examples of what is wrong with junk websites that masquerade as "Web Services". MSN, Yahoo, etc. are junk sites.

Google's main websites - search, images, groups, news, etc. are so extremely useful I don't know what I would do without them. Slashdot and Freshmeat and Source Forge too all have their usefulness. But they are quite specialized.

Here is how I would implement the Free Software Network.

  1. Start Small
  2. No Advertising
  3. User Created Content
  4. Work With Other Networks
  5. Expand beyond just the Web

Start Small

Section the Website like a newspaper - News, OpEd, Arts, etc. Supply articles and feedback. Have some forums. Have classifieds.

But -- I would not design the site to be as people's single source for Free Software information. This is the largest fault of so many sites (and magazines and newspapers). It's like CNN having Weather and Sports reports -- what a waste. Perhaps some people like the all-in-one approach, I don't. CNN's Weather and Sports reporting are lame because they just do not -- can not -- spend enough time and effort in producing them. I lean toward the maxim, "Do one thing, and do it well." But, one can do several things well. Yet, one can not do all things well.

Grow as needed, be prepared to cut back if needed. I like what Ray Bradbury says, "The writer learns what to leave out."

No Advertising

When I want to buy a stereo, I will go looking. I certainly will not rely on waiting until I see an advertisement before looking to buy. Perhaps it is just me -- I never respond to advertising except when I am actively looking for something.

Here is how to make money without traditional advertising (banner ads and random/rotating, unrelated animated graphics):

Create a sections for the discussing of solutions provided by those Free Software related companies, organizations, and groups that will be willing to pay -- either by direct dollars, equipment or services etc. These "sponsors" then become participants, or partners (although not controlling partners, they would need to have no say in final web content) whereby they would offer press releases, new product notifications, etc. The FSN would supply the sponsors a place where people can come and discuss their products and services, freeing the sponsors from creating this kind of network on their own. The FSN would supply the sponsors reviews and articles and feedback.

I am sure that companies and organizations and groups would pay for these services.

(Also, one would ask for -- but not require -- donations.)

User Created Content

FSN users would have a "section" of the Network for their own use, modeled after Indymedia.

Work With Other Networks

At first, in small steps, link up to sites like Wikipedia, Creative Commons, and other truly (100%) Free Websites and Networks. Adhere to stringent guidelines.

Expand beyond just the Web

And this is the best and most important part: Expand to publishing: a Music Label, a Magazine (think 2600 and NOT PC World), Books.

Then, somehow either work with existing Free Software-like companies such as Web Design, and Graphics Design. Again, with strict guidelines about Free as in Freedom -- i.e. pay for services like creating the design and not the design itself.

Well, that is my vision of a Free Software Network.

what else?, posted 4 Oct 2005 at 22:57 UTC by lkcl » (Master)

prozac, thank you.

what else: i believe that the ability to create and edit work (and print) wherever you are is also important.

i am _delighted_ with the concept of nomachine (nxserver / freenx): free software finally gets something equivalent to RDP.

running openoffice etc. on centralised servers with NX clients accessing them from anywhere in the world is the "clunky" first version.

the second version is to be able to sync documents back-and-forth from the user's machine (if they in fact have one) such that local editing, printing and other access can be done without significant delays.

again - a "clunky" way to achieve that is to constantly have openoffice doing autosave in the background, and then constantly rsync'ing the document to the workstation.

the _ultimate_ version: in a peer-to-peer network, not knowing _where_ openoffice is actually running :)

re-read, posted 4 Oct 2005 at 22:59 UTC by lkcl » (Master)

i just read your reply again. i like it.

a newer impoverished version of Pablo Neruda's poem 'disaction', posted 8 Oct 2005 at 15:38 UTC by sye » (Journeyer)

DISACTION

The dove is filled with spilt papers its breast is stained with erasers and weeks, with blotting computer whiter than a corpse and screens frightened by their sinister color.

Come with me to the shadow of internetworkings tothe weak, delicate, pallid color of the chiefs, to the tunnels deep as calendars, to the doleful thousand-mega disk.

Let us examine now the licenses and the conditions, the special affidavits, the vigils, the petitions with their teeth of nauseous autumn, the fury of ashen destinies and sad decisions.

It is a tale of wounded bones, bitter circumstances and interminable languages, and opensource suddenly serious.

It is the dead of night, the veinless head from which day suddenly falls as if from a bottle broken by a lighteningbolt.

They are the feet and the clocks and the fingers and a locomotive of dying enterprise, and a bitter sky of soaked silicon, and a yellow river of smileys.

Everything reaches the tips of fingers like flowers, and fingernails like lighteningbolts, withered keyboards, everything reaches the ink of death and the violet mouths of the copyrights.

Let us weep for the death of earth and fire, swords, grapes, the sexes with their tough realms of roots, the alcohol ships sailing among ships and the perfume that dances at night, on its kneees, dragging behind a planet of perforated roses.

With safari's suits and stains on our browsers let us fall into the depths of google searches, into the anger of enchained keywords, into demonstrations tenaciously defunct, into systems wrapped in blue ciscos.

Come with me to the Offices, to the uncertain smell of ministries, and tombs, and postage stamps. Come with me to the white day that is dying screaming like a murdered bride.

the price of freedom, posted 8 Oct 2005 at 21:34 UTC by lkcl » (Master)

it became a nightmare above all else, enriched with cinders of an ashen dream. the power of freedom asks a heavy price: our intelligence is not all it may seem...

hehe, posted 12 Oct 2005 at 15:31 UTC by exa » (Master)

can you spell troll?

GNU GENERATION 2005

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!

X
Share this page