Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : LAST MILE TECHNOLOGIES - Let's Discuss Them Here -- Ignore unavailable to you. Want to Upgrade?

To: Raymond Duray who wrote (4603)	7/13/1999 1:19:00 PM
From: WTC	Read Replies (1) \| Respond to of 12823

The ISDN arrangement described can certainly be made to work, but you correctly surmise that it is complicated and can be error-prone. I suggest the best indicator of success would be your impression of the LECs desire to get it installed and operating right. I have no experience with your LEC contacts, but from what you quote, it seems to me you would be showing a lot of optimism to expect quick and competent installation and maintenance. Sounds like these guys are willing to live up to what they consider the letter of the agreement with the PUC to provide ubiquitous ISEN service, but they will do their damnedest to talk a customer out of placing an order. Remember that just getting the order written correctly for ISDN so that the switch optioning can be done properly can be daunting -- especially if there is little ISDN service order experience or motivation to get it right. Also, the same guy who judged the noise on your line by listening to his Budd set and alluding to high-open splices might be your ISDN installer. Does that give you a warm feeling? Your instincts on this seem good to me. The $500 is outrageous, but it may well be correct, based on the PUC negotiations. Obviously, I just don't know. You might want to double check the actual tariff rates, including adders, that could apply to initiate your service. You might also want to think of the ISDN service as 128kb/s, to give yourself the motivation to explore this any further at all.

To: Raymond Duray who wrote (4603)	7/13/1999 1:30:00 PM
From: WTC	Read Replies (1) \| Respond to of 12823

I am curious -- why no comment here on that newly-minted Last-mile darling -- Metricom (MCOM)? Languishing between 5 and 7 for months, a deal gets done, and we don't see a doubling in share price, but a 6x move! Now it bounces around a bit below that, but it still makes the NASDAQ most active lists consistently after being thinly traded for months. I have been conflicted out of commenting prior to the recent public announcement, but even with some non-public info, the meteoric rise caught me off guard. Anyone see implications for Metricom clones? Is this the market showing a lack of confidence in 3G wireless (or IMT-2000, if you will)? How close does another company have to look like the Metricom network topology and business plan to climb on this rocket?

To: Raymond Duray who wrote (4603)	7/13/1999 1:34:00 PM
From: lml	Read Replies (1) \| Respond to of 12823

Ray: I can relate to your situation. I'm 38K ft out. I presently use ISDN to connect to the Internet that has a repeater located at a Pair Gain box, which then feeds to the CO via fiber. I hope to get ADSL from PacBell by year-end via installation of 3GDLCs from AFCI. See biz.yahoo.com Typically, an ISDN pair is provisioned for 2 B channels each delivering 64K bps of bandwidth, which can be multi-linked together to deliver 128K bps. However, unless you can bypass your serial port with an accelerator card or connection through another port, the multiplexed bandwidth is limited to 115.2K bps. So the bandwidth comparison should be about 50-53K bps over standard lines v. roughly 100-110k bps over a multiplexed ISDN line. There is a difference when surfing the net, but it is minimal. A more material difference is experienced with file transfers. The upfront costs you speak of, which I assume includes the modem, is a bit high, but then again, you're in Oregon with USW & not in Calif with PacBell or GTE. Given the broadband "rush," you may opt to hold off purchasing ISDN. You might want to look into satellite via DirecPC, though this will only enhance your maximum downstream bandwidth to roughly 250-350K bps. Alternatively, you might consider if "Shotgun" technology is offered by any ISPs in your area. This technology allows the multi-linking of 2 separate POTS lines each delivering 56K bps, that result in 112K bps of bandwidth. In order to achieve this, your ISP will have to be provisioned with the same technology as the subscriber's "Shotgun" modem, & you will likely pay a higher fee since the subscriber would connecting to 2 ISP routers instead of just 1 router. Hope I've shed some light.

To: Raymond Duray who wrote (4603)	7/13/1999 3:51:00 PM
From: John Stichnoth	Read Replies (1) \| Respond to of 12823

Ray, I had ISDN on a trial from Bell Atlantic last year. I quickly cancelled because I wasn't getting the benefit I wanted. Problems included: 1. I had to dial into a separate, different number from the 28.8 modem. The ISDN line was frequently busy. 2. They used a separate server for the ISDN, apparently, so that when I did get through I was in a separate bottleneck that was worse than the servers on the 28.8 side. 3. The response time that I experienced wasn't all that impressive. Most of the time I was accessing only one line (the 64.4 you mentioned), so not much benefit. When the second line would kick in--which it didn't do very often--the whole thing would seem to take a deep breath before running at the higher speed. Since the larger files I was downloading only occurred intermittently, that "deep breath" actually seemed to slow things down a lot of the time. 4. The use of the ISDN line was metered, so in the back of my mind I was trying to save money by getting off fast. Not very much money, but distracting. 5. To add insult to injury, the call to the ISDN server was actually a toll call! A local toll call (I think the amount for the month was a couple of dollars), but insulting after all the other shortcomings. I haven't checked BEL's offering lately. Above offered as to some points to watch out for if you are considering it.

To: Raymond Duray who wrote (4603)	7/14/1999 12:59:00 AM
From: Raymond Duray	Read Replies (1) \| Respond to of 12823

Re: Bruneau on Bandwidth Glut! SIer responds, "Oh, really?" Threaders: I found this response on the LU thread to the Sunday Boston Globe article to be very perceptive and articulate: Message 10464890 To: stephen karasick (8553 ) From: Jack Whitley Monday, Jul 12 1999 12:21AM ET Reply # of 8584 <<Telecom industry may be left severely overbuilt after rush to add 'bandwidth' capacity By Peter J. Howe, Globe Staff, 07/10/99 f Mark Bruneau is right, tens of billions of dollars worth of stock market investment in information-highway communications networks runs the risk of going up in smoke.>> Ludicrous. This person doesn't really include bandwidth multipliers in his "forecast" of future bandwidth demand. 1) On the web site I help administer, user session traffic is up 4-fold in the last nine months to six figures, with bytes requested per month up 16-fold during the same nine months. And we have only 1 per cent of our imaged inventory on the site in the form of compressed 2D JPEGs. How much more product could we sell if we could send a short video of each of 120,000 items, on request ? How much additional bandwidth will that require, times the millions of web businesses out there with digital assets. Does a software company have a better chance to sell its product via a net download of 25 minutes, or 10 seconds? Can we have too much bandwidth ?? 2) SSL/encryption - what percent of total public Internet traffic is now encrypted? Very little. How much will be encrypted ? Hmmmmmm. E-commerce, VPNs, etc., I would venture to say large amounts. SSL transactions vs. non-SSL transactions currently degrade Apache server capacity by a factor of 50 or more, and definitely mean exponentially more packets coursing the fiber in the future. 3) The dreaded 30 fps video on demand. 4) As exponentially faster edge connections for users are rolled out to the entire net surfing population, exponentially more throughput will be needed in the core. 5) VoIP - will this really work unless there is massive over-provisioning of bandwidth? It would be impossible to produce non-latent voice in an environment rife with packet collisions, which is what we have now (and there are very few IP voice packets sent currently on the web). 6) Price/demand elasticity - cheaper bandwidth available in larger quantities ? Mr. Market says cheaper, faster bandwidth means more users, more packets requested/sent, more bandwidth needed. 7) One computer per 100 people on earth. Do you think there is a chance that this ratio may go up :~) Multiply all of the above multipliers by this multiplier. Can the transactional processes of the web ever be made too fast for business? Will we be there two years from now ? Am I selling my LU stock based on Mr. Bruneau's assesment? No. jww