SI
SI
discoversearch

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: WTC who wrote (2573)12/14/1998 2:15:00 PM
From: Frank A. Coluccio  Read Replies (2) | Respond to of 12823
 
Bill, another well though-out reply, thanks. Was my tongue pressing the wall? Yes, in a way, but with an extreme amount, let me repeat that, with an extreme amount, of reverse tension being applied.

Let me see how much of your message I can reply to while I have some time left for lunch. Don't mind me if I type with food in my mouth.

Notice is Hereby Given to Onviewers: This is only a high-level academic discussion!
--------

FAC: >Perhaps there should be four franchising privileges in every locale for the ILEC, the CableCo, the utility company, and a fourth one for all others.<

WTC: >>Why limit wireline access competitors (CLECs) as long as there is no inherent limitation like spectrum?<<

If you view the totality of my upstream message, this suggestion was only one of several options, and in retrospect it could have been lumped into several permutations with the others. But we'll take it as a standalone for now, for discussion purposes. And that's all this is, an academic discussion, lest anyone conclude differently.
----

Spectrum is a scarce resource, according to many [and some disagree, depending on their modulation choice du jour], and that is why it's rationed, allocated and ultimately regulated. And I'm sure there are some more sinister reasons that some may offer up here, but I'd rather not get into those now.

Pole space and interconnection attachments, and real estate in the pedestal, all fit into this category, as well.

The potential confusion caused by dozens if not hundreds of aspiring CLECs, all contending for interconnection in the field... where the unbundled wire elements, or UNEs, would be broken out, is great, and I would suggest in many parts, untenable. One could envision a single household having live connections in place to three or five different providers, one to suit a different specialization in the future. Which isn't all bad, just unwieldy, to the point of impractical.

[I can see the last mile access aggregators lining up at the door now.]

You've made this last point yourself in an earlier post, which is what started me off on this hunt in the first place.

I'd rather not refer back to photos which by now are cliche, depicting the maze of pole attachments that once mired the 'landscape' of Wall Street during the turn of the century. But these visions are not beyond contemplation when it comes to the present-day situation we are discussing at the UNE in many regions and locales, and certainly in suburban and urban areas.

Micro miniaturization can, in fact, be used to ameliorate the potential gravity of this situation, say, by the use of digital routing and cross connect assemblies, but said components would be neutral in nature, leaving the unanswered question as to who would pay for them? And to take this approach at the routing level, then, would be no different than to unbundle the upper layers as I've offered in another option, at the central office or deeper into the cloud.

This is where the buildout for "all others" option comes in.

If such a fourth buildout were permitted by all non-ILEC contenders under the auspices of a consortium umbrella effort, then it's at least plausible that these answers could get answered. I know that this would be a Manhattan Project-like undertaking, but if that's what it takes to win a war, then so be it.

The fourth build out would not represent hardly as much of an egregious environmental impact, or administrative boondoggle with a recalcitrant ILEC or CableCo MSO, as would be the case if the multitudes were allowed to pull cable and dig up controlled environmental vaults (CEVs) one by one. And to do a one-by-one deployment of physical latey, would cause each player to price themselves out of business before they could ever got started.

Rather, it would be a single endeavor at the physical layer (hopefully all fiber, with copper coax and t.p. pigtails being limited to rare cases) using fiber as a least common denominator, potentially to the residence or business location as well, with neighborhood caging for up to a certain amount of competitors [here we go again, how many would a certain amount be?] which could be overseen by quasi-municpal entities, or professionally run colo businesses, a la re-engineered carrier hotels.

The Q-M or colo entities' charters would include Chamber of Commerce-like verbiage in every case. The full service area network (FSAN) architecture would do nicely here, but that would only be one economy-size options for the competitors. Other overlays would be permitted on the basis of space division, or routing, or whatever.

WTC: >>Isn't a taxi medallion, or an IBM reseller medallion of old, about as anti-competitive as it gets? Aren't these designed quite specifically to protect incumbants from the downward price pressure that comes from unrestrained new competitive entrants? Where is the public good there?<<

In a way, that's correct. The medallions, along with testing and licensing, are also designed to ensure that some minimum qualifications are met prior to allowing entry (or at least that was the initial intend... please, no cabby jokes here...;-)

But my point about minimum qualifications should not be lost in this discussion. Every future connection on the 'net, whether physical or virtual, will have the potential of affecting service to every other point on the 'net... dependent on where a call is initiated from and to. We'll leave carrier certification, however, to another thread.

WTC: >>I think you are mixing baseline situations and reaching a highly suspect conclusion (I'm assuming here that your tongue is not pushed hard into the side of your face.)<<

FAC: Like I said, reverse tension was used.

WTC: >>Your examples are all based on limited allocation of spectrum -- the FCC needed some way to fit 'x' prospective service providers each needing 'y' MHz of spectrum into 'n' MHz of spectrum available. We might argue that the results from comparative hearings, lotteries, and some auctions are so ugly that we certainly don't want to go there unless there is no choice whatever. There certainly seem to be choices with wireline.<<

FAC: What are they? Assuming that unlimited access is permitted, of course. On that score, if there were only a handful of fools, er... competitors, who would want to get into this race, I too would see no problems except for the artificial barriers that have already been erected by the incumbents.

WTC: >>This is a cogent argument (perhaps unintentional) against some interesting ideas that have been put forward by Diamond Lane proposing what they term "Logical Collocation" -- what has been termed elsewhere in discussions "virtual collocation." ...Under that scheme, the ILEC would sell and deliver a UNE or a combination of "glued" UNEs that incorporate ILEC subloops, ILEC DSLAMs, ILEC ATU-R and ATU-C, and ILEC fiber backhaul facilities from the remote DSLAM to a CO hand-off point, say, the MDF. Those ideas have a lot of merit for sidestepping the snarl of collocation complexity in field locations, but they create some new requirements for new shared access DSL management systems that all participating CLECs can access, with the ILEC presumably managing the (OSS)system. <<

FAC: We were good up until that last sentence. In my scenario, the OSSs would be the purview of a company or other form of entity whose "business" was solely that of colocation. There are a growing number of such firms, who I am sure would be more than happy to take on such an undertaking on a cost-plus basis.

WTC: >>Such a system does not presently exist, but it would not seem like a man-to-the-moon level development effort.<<

What'e the alternative? We've waited three years come next month to see the fruits of regulated deregulation, and nada. Actually, while we're discussing this, I think that LVLT is already onto this opportunity, and they are already proceeding with a colo model [that combines at least two of my stated options] that may be a good fit here, although their means of transport may differ radically from what some of the CLECs have in mind.

WTC: >> The bigger criticism of such a plan has to be that it could discourage innovation and delivery of new services via new DSL technology that is not part of the ILECs "approved list" of compatible and supported vendors/models.<<

FAC: BINGO!

WTC: >>This problem hurts everyone: the customers could be denied the benefits of innovation; the CLECs could be denied opportunities for differentiation in the market; <<

Not under my neighborhood caging scheme

WTC: >>...and the ILECs descend into a purgatory where everyone is mad at them even though they are working very hard in good faith to deliver what they promised, probably successfully at some point.<<

FAC: I won't argue your assessment, except to state that it may be true over time, because it would require prescience on my part. But there is something intrinsically wrong with a model that has the hens continuously being cared for, and nurtured, and monitored, by the fox. And that is precisely what we would have here under a UNE model, structural separations or no structural separations, IMO.

WTC: >>The requirements would certainly be constantly in flux, with new technological capablities needing to be added to the "supported" list faster that the ILECs can assure themselves the new items work as promised, and don't interfere with the rest of the installed "approved" hardware and software. <<

Such would be the case if we followed the road to ILEC reuse and collocation. It need not, however, be the case in a totally partitioned fourth buildout scenario.

Lunch is now over. I'll look forward to replying to the remainder of your message later on tonight. In the meantime, comments and corrections are always welcome.

Regards, Frank Coluccio



To: WTC who wrote (2573)12/14/1998 8:26:00 PM
From: Frank A. Coluccio  Read Replies (1) | Respond to of 12823
 
To continue from this afternoon,

>>The g.lite standards and products that come to market in late 1999/ early 2000....It will not (at least initially, and probably for a long time) support the features that many businesses are looking for: managed data services, QOS guarantees, high guaranteed upstream data rates, etc. So this is really not a solution for the market space where virtually all the CLEC DSL activity is now -- in small and medium business. These customers are mostly buying 384kb/s symmetric SDSL services from their DLECs, and it is hard to see where any announced collaborations really address this market space. <<

Agreed. Have you checked out some of the rates the data CLECs are asking for fractional T speeds? They're higher in some cases than those for ILEC T1s [okay, there are other things at the ISP level being thrown in at times, but still... the rates they are asking are way high, IMO.

>>I won't try to explain thought processes at even one ILEC, but I would suggest that it has to be a mistake to try to characterize thinking at multiple ILECs monolithically.<<

Touche! Old habits don't die quickly. But you are right.

>>... and BLS and BEL seem from my perspective to be trying to make DSL services roll outs work. The perhaps fair criticism of BLS and BEL is that they might be trying to do too much for DLECs in the loop qualification arena, when the DLEC just wants a loop and he will see if he can make it work.<<

That's an interesting insight that I had not heard before. Thanks.

>>You followed with a great anecdote that no one who has been near where you worked would doubt for a moment. Maybe we are lucky that the DSL provisioning process is handled out of exchange services, rather than special services, at least in BEL. Those special services guys, well, you captured it well.<<

I see that you too have had the pleasure.

Regards, Frank C.
 
ps - do you notice how I kinda mellow out after dinner? ;-)



To: WTC who wrote (2573)12/15/1998 7:51:00 PM
From: Frank A. Coluccio  Respond to of 12823
 
WTC, and All, re: Virtual Unbundling [(c) 1998, here in SI LM ]

In an earlier message here, I suggested that unbundling could be achieved at the upper layers, instead of at the physical layer unbundled network elements, or UNEs. It should come as no surprise, then, that I think that this is the way of the future. We are seeing more evidence of it every day. To differentiate this from physical unbundling, I call it virtual unbundling. Quaint, but effective.

We already have such a form of unbundling now, to be sure, for this accounts for how we access information on the Internet, and beyond, under the umbrella of IP routing. But we don't regard this as unbundling in the same way that we think of gaining access to other forms of more traditional services, which have their roots in more established forms of nailed up lines. For example, POTS voice, fax, point-to-point dedicated data lines, etc.

Instead, we tend to think of unbundling as gaining access to discrete physical elements that ILECs would break out for sale or lease to a quasi-facilities based competitor, or CLEC, for the purpose of making those same rudimentary nailed up services available to end users under a different brand name.

But we ordinarily don't think of it as "unbundled" when these same services are enabled when they are made available over ATM and IP-governed fabrics, yet. This is the distinction I have been pointing to, for I believe that this is the way of the future... where service providers of varying types will provide access to customers for a wide range of customary-like and new services over a common pipe, which in fact could be the actual facility pipe provided by any one of the available facilities-based providers. It won't make a difference, much, except for cable tv delivery products, which of course, will still need special handling. Right? Come to think of it...
---

We spoke about this about 18 months ago, or so, here in the Last Mile thread briefly with Ray J., and then later in the VoIP thread. Within the course of the past six to eight months, we've seen some of the larger players enter this realm, inconspicuously, in the sense that it doesn't appear that they ae doing anything out of the ordinary, only something a little askew of Internet access. Shrewd.

For example, Sprint appears on the scene and they announce that they will be rolling out ION, an all-in-one solution for voice, data and Internet access that will be supported primarily over IP-over-ATM-over-DSL-over-Copper.

T would do the same with INC at some point for corporate branch office customers.

And MCI with their end to end direct connect, whatever they call it.

T is going one step further, in that they now intend doing this over the last mile over CableTV facilities as well, for residential users. I suspect that T will very likely hold tight reigns on what is "permitted" into to their cloud space, if they can find a way to achieve this without compromising the basic spirit and "intent," whatever that is turning out to be, of the Internet.

I say they can't do that, and if they find a technical solution to doing this, they will face regulatory problems, and lose share at the same time. If they want to play IP, then they'll have to open it up. And this is one of the reasons why I thought that all of the hullabaloo that T put up about not wanting to unbundle their new-found cable plant was a farce. Once they open up to the Internet, they will have opened the gates to choice, anyway.
---

I just came across an encouraging press release over in the LVLT thread, posted by Kevin McNeill, which discusses a consortium of some heavy hitters who not only have the same idea, but they are now doing something about it at the standards level.

A few posts earlier upstream in this discussion, I stated that LVLT was one of those who would be at the front of the pack to do the colo thing, suggesting that they would be one of the first players to make themselves available to substitute for the traditional ILEC function. LVLT and some of the others in the mix, below, are going to be the front runners in this new model at the virtual level as well as, in some cases, facilities based as well, I feel.

IMO, even though this release is not a direct hit on the space I'm referring to, it's close enough to suggest that the stage is now set for this very kind of virtual unbundling to eventuality unfold. And once they have the solution-set [protocols] to proceed, they will ride the wave high on top of other peoples' physical facilities, and it will neither make a difference who provides the physical link, nor take a Manhattan Project-like effort to get it going.

It will only require that an end user have in place the right client software [probably embedded in a household interface unit with appropriate I/O extensions and outlets, so as to make it as friendly and transparent as possible] and a decent basic access line to the Internet, although the higher the speed, obviously, the better. Not bad.

What do you think about this new trend to unbundle at the higher layers? Anybody?

Enjoy the press release below, and Regards,

Frank Coluccio

---
exchange2000.com

biz.yahoo.com
---

Tuesday December 15, 11:24 am Eastern Time

Company Press Release

SOURCE: Level 3 Communications

Telecommunications Carrier Group Forms

Coalition Advocates Rapid Adoption of Protocol Standards to
Integrate the PSTN with Newer Internet Protocol (IP) Networks

OMAHA, Neb., Dec. 15 /PRNewswire/ -- A coalition of telecommunications carriers has announced the formation of the Packet Multimedia Carrier Coalition dedicated to the rapid establishment of protocols designed to bridge between current circuit-based public switched telephone networks (PSTN) and emerging Internet Protocol (IP) technology based networks.

Coalition members, including Frontier Communications, GTE, ICG Telecom Group, Illuminet, IXC Communications, Level 3 Communications, NEXTLINK Communications, NTT America, SBC Technology Resources, Sprint, TESS, LLC, Time Warner Telecom and Williams, are committed to providing customers with the best of both the traditional PSTN and the newer IP technology networks. To do this, new protocols are needed that will allow the seamless integration of these two types of networks. Such integration will enable customers to benefit from the lower cost of IP network services, including voice and fax, without modifying existing telephones and fax equipment or dialing access codes.

The goals of the Coalition include development of a number of protocol sets that address issues related to the bridging of thePSTN with IP technology networks. For example, one protocol set would outline requirements for traffic exchange between the PSTN and IP networks in such areas as interoperability, reliability and performance. Once developed, the protocols would be submitted to both the Internet Engineering Task Force (IETF) and the International Telecommunications Union (ITU) standards bodies.

''The Packet Multimedia Carrier Coalition gives carriers a clear voice in the standards making process,'' said Isaac Elliott,chairman of the Coalition and senior director of voice network engineering for Level 3 Communications. ''Our aim is to provide input into the standards approval process at both the IETF and the ITU.''

''This Coalition was established to facilitate the rapid deployment of the kinds of new technologies our customers are demanding,'' said Mark Hewitt, senior director of engineering and product development at Frontier Communications. ''We fully support its goals.''

''We believe voice over Internet Protocol, as well as other Internet Protocol technologies, can be enablers for making information available to any customer with any type of communication device,'' said Robert Flood, ICG chief technology officer. ''This will, however, require a more complete suite of standards and signaling functionality to tie the existing telephony infrastructure to a future packet-based infrastructure. We see the Packet Multimedia Carrier Coalition as an important vehicle to achieve these goals.''

''The significance of the Coalition is its charter of interoperability not only between IP carriers but also the PSTN,'' said Lori Mullane, manager of business development at Illuminet. ''It is imperative that regardless of the underlying infrastructure that is utilized by a carrier -- whether it is circuit-based or packet-based --- that service remains seamless to the customer. An open carrier forum helps achieve this.''

''The Carrier Coalition provides a means for all communications service providers to address interoperability issues,'' said Chris Rothlis, vice president of new product development at IXC Communications. ''Customers are the ultimate beneficiaries of the Coalition's efforts. Through efforts like this, communications service providers will deliver on the promise of converged networks.''

''The advance of standards providing for the interoperability between legacy networks and equipment and next generation networks is essential to the future of our industry,'' said Doug Carter, senior vice president of technology at NEXTLINK
Communications. ''We look forward to the point when these standards exist, allowing for the construction of networks based
on open, cost-effective, standards-based equipment from a broad array of participating vendors.''

''NTT America supports the goals of the Coalition which include providing a unified voice in the development of telecommunications protocols,'' said Kaz Yozawa, vice president of NTT America. ''Now is the time to look at standardization with the emergence of the Internet Protocol technology based networks.''

''Sprint is very actively involved in the development and near term deployment of broadband network service delivery elements that can be advantaged by uniformly designed control interfaces,'' said Terry Yake, vice president of applied research at Sprint Long Distance Division. ''Standardized interfaces will give greater opportunities for service providers to select among the best manufacturers of control and inter-working devices.''

''TESS is looking forward to the new standard being defined by these efforts,'' said Stephan H. Andrade, president of TESS. ''The true beneficiary will be the customer by having a better network.''

''Time Warner Telecom looks forward to participating in the interoperability of IP and PSTN,'' said Graham Powers, senior vice president, engineering and technology. ''We are pleased to be part of the Packet Multimedia Carrier Coalition team and believe it will be instrumental in bringing voice over IP services to the marketplace.''

Carriers Support MGCP

In the area of external control of media gateways, the Coalition announced its intention to support the IETF media gateway control standard when finalized.

''We are encouraged and excited by the Media Gateway Control Protocol (MGCP) draft that has been submitted to the IETF mailing lists,'' said Michael Vent, executive vice president and chief technology officer at IXC Communications. ''We are seeing rapid progress toward an acceptable standard in the IETF and recognize standards as a key element of future network quality and interoperability.''

MGCP represents a combination of the Internet Protocol Device Control (IPDC) specification developed by a consortium made up of leading communications hardware and software companies and the Simple Gateway Control Protocol (SGCP).

MGCP enables external control and management of data communications equipment operating at the edge of emerging multi-service packet networks -- known as ''media gateways'' -- by software programs known as ''call agents'' or ''media gateway controllers.'' Examples of media gateway devices include voice over IP gateways, voice over ATM gateways, modem banks, cable modems and set-top boxes, soft PBXs, and circuit cross connects.

A draft of the MGCP specification was recently submitted to the IETF as well as the European Telecommunications Standards Institute Telecommunications & Internet Protocol Harmonization Over Networks (ETSI TIPHON) working group for reviewand modification. The draft can be found at level3.com.

About the Packet Multimedia Carrier Coalition

The Packet Multimedia Carrier Coalition is an open body and welcomes participation by communications service providers. There are no fees associated with Coalition membership. For more information on the Coalition, contact Level 3 Communications via email at Carrier.Coalition@Level3.com. Member companies include:

Frontier Communications, a unit of Frontier Corporation (NYSE: FRO - news), is one of the leading providers of integrated communications services -- including Internet, IP and data applications, long distance, local telephone and wireless -- to business customers nationwide. The self-healing Frontier Optronics Network(SM) provides customers with faster transmission speeds, greater bandwidth capacity and unrivaled reliability. For more information, visit the Frontier web site at www.frontiercorp.com.

With 1997 revenues of more than $23 billion, GTE is one of the world's largest telecommunications companies and a leadingprovider of integrated telecommunications services. In the United States, GTE provides local service in 28 states and wireless service in 17 states; nationwide long-distance and internetworking services ranging from dial-up Internet access for residential and small-business consumers to Web-based applications for Fortune 500 companies, as well as video service in selected markets.

ICG Netcom is the brand name for products and services from ICG Communications, Inc. (Nasdaq: ICGX - news) and ICG's subsidiary, NETCOM On-Line Communication Services, Inc. The combined entity is a leading integrated communications provider (ICP) offering high-quality telecommunications services. Headquartered in Englewood, Colo., ICG has extensive switched fiber-optic networks and offers local, long distance and enhanced telephony and data services in California, Colorado, the Ohio Valley and parts of the southeastern United States. The company provides Internet communication solutions, connectivity and Web site hosting to individuals and to small- and medium-sized businesses through its subsidiary, NETCOM On-Line Communication Services, Inc. ICG also is a leading designer and installer of copper, fiber and wireless infrastructure for buildings and campuses. Further information is available on ICG's Web site located at icgcomm.com.

Illuminet (www.illuminetSS7.com) provides advanced signaling network technology and enhanced applications, including local number portability, wireless fraud management, SS7 network usage measurement and billing services to more than 1,000 local, interexchange, wireless and multimedia carriers throughout North America and in parts of Asia.

IXC Communications' (Nasdaq: IIXC - news) network-based delivery solutions are designed to address the speed and capacity requirements of the global communications market. Having completed the first new coast-to-coast fiber optic network in the United States in more than a decade, IXC Communications, Inc. is at the forefront of the industry's new class of emerging domestic and international carriers. IXC offerings include private line, broadband, Internet and long distance switched and dedicated services. IXC is a publicly traded company listed on Nasdaq under the symbol IIXC. IXC's Web site is at www.ixc-comm.com.

Level 3 Communications, Inc., is a communications and information services company that is building the first international network optimized for Internet Protocol technology. The Level 3 Network will combine both local and long distance networks, connecting customers end-to-end across the U.S. and in Europe and Asia. The company expects to complete the U.S. inter-city portion of the network during the first quarter of 2001. In the interim, Level 3 has signed an agreement to lease a national network over which it began to offer services in the third quarter of 1998. Level 3 will provide a full range of communications services (including local, long distance, international and Internet services). Level 3's common stock is traded on The Nasdaq National Market under the symbol LVLT. Its World Wide Web address is www.Level3.com.

NEXTLINK Communications, Inc. was founded by Craig McCaw in 1994 to provide local, long distance and enhanced communications services to commercial customers and is one of the fastest growing competitive telecommunications providers in the nation. Headquartered in Bellevue, Wash., NEXTLINK currently operates 19 facilities-based networks providing switched local and long distance services in 33 markets in 11 states.

NTT America was founded in 1987 as a wholly-owned subsidiary of Nippon Telegraph and Telephone (NTT), one of the world's largest telecommunications service providers. Today, it is recognized as a leading provider of state-of-the-art products and services, as well as a major player in the quest for new communications solutions through innovative research. Its World Wide Web address is www.nttamerica.com.

SBC Technology Resources, Inc. (TRI) is the applied research subsidiary of SBC Communications Inc., the $24 billion communications leader formerly known as Southwestern Bell Corporation. TRI provides technology consulting and expertise to the SBC Communications family of companies, exploring new ways to incorporate leading-edge technology into communications products and services.

Sprint is a global communications company -- at the forefront in integrating long distance, local and wireless communications services and the world's largest carrier of Internet traffic. Sprint built and operates the United States' only nationwide all-digital, fiber-optic network and is the leader in advanced data communication services. Sprint has $14 billion in annual revenues and serves more than 16 million business and residential customers.

TESS, LLC. is a private Integrated Communications Provider, offering Company offering innovative, cost-effective communications solutions for small to medium-sized businesses and residential customers. Headquartered in Evergreen, Colo., TESS is working in the southwestern United States offering local dial-tone, long-distance, internet and video services.

Time Warner Telecom, headquartered in Greenwood Village, Colo., is a fiber, facilities-based integrated communications provider of local telephony products and services to medium and large business customers in 19 U.S. cities.

Williams (NYSE: WMB - news), through its subsidiaries, provides a full range of traditional and leading-edge communications and energy services, and is the nation's largest volume transporter of natural gas. Williams' communications enterprises include
U.S.-based wholesale multimedia fiber-optic network; North American single-source business communications systems integration; international video satellite and fiber-optic transmission, multipoint video and audio conferencing and satellite business applications. Williams information is available at www.twc.com and www.wilcom.com.

The statements made in this press release may be forward-looking in nature. Actual results may differ materially from those projected in forward- looking statements. Additional information concerning factors that may affect actual results and other important factors can be found within the filings with the Securities and Exchange Commission made by each of the members of the coalition. Statements in this release should be evaluated in light of these important factors.

SOURCE: Level 3 Communications